i
“Monitoring and control of Delamination in Drilling of
GFRP (Glass Fibre Reinforced Plastics)”
A Thesis submitted to Gujarat Technological University
For the Award of
Doctor of Philosophy
in
Mechanical Engineering
by
Patel Jaykumar Bipinbhai
Enrolment no.119997119010
Under supervision of
Dr. M.B.Patel
GUJARAT TECHNOLOGICAL UNIVERSITY
AHMEDABAD
March - 2019
ii
© Jaykumar Bipinbhai Patel
iii
DECLARATION
I declare that the thesis entitled “Monitoring and control of Delamination in Drilling of
GFRP (Glass Fibre Reinforced Plastics)” submitted by me for the degree of Doctor of
Philosophy is the record of research work carried out by me during the period from July 2011
to December 2018 under the supervision of Dr.M.B.Patel and this has not formed the basis
for the award of any degree, diploma, associateship, fellowship, titles in this or any other
University or other institution of higher learning.
I further declare that the material obtained from other sources has been duly acknowledged in
the thesis. I shall be solely responsible for any plagiarism or other irregularities, if noticed in
the thesis.
Signature of the Research Scholar: …………………………… Date: ….………………
Name of Research Scholar: Patel Jaykumar Bipinbhai
Place: Ahmedabad
iv
CERTIFICATE
I certify that the work incorporated in the thesis “Monitoring and control of Delamination
in Drilling of GFRP (Glass Fibre Reinforced Plastics)” submitted by Shri Patel
Jaykumar Bipinbhai was carried out by the candidate under my supervision/guidance. To
the best of my knowledge: (i) the candidate has not submitted the same research work to any
other institution for any degree/diploma, Associateship, Fellowship or other similar titles (ii)
the thesis submitted is a record of original research work done by the Research Scholar
during the period of study under my supervision, and (iii) the thesis represents independent
research work on the part of the Research Scholar.
Signature of Supervisor: ……………………………… Date: ………………
Name of Supervisor: Dr. M.B.Patel
Place: Ahmedabad
v
Course-work Completion Certificate
This is to certify that Mr. Patel Jaykumar Bipnbhai enrolment no. 119997119010 is a PhD
scholar enrolled for PhD program in the branch Mechanical Engineering of Gujarat
Technological University, Ahmedabad
(Please tick the relevant option(s))
He/She has been exempted from the course-work (successfully completed during
M.Phil Course)
He/She has been exempted from Research Methodology Course only (successfully
completed during M.Phil Course)
He/She has successfully completed the PhD course work for the partial requirement
for the award of PhD Degree. His/ Her performance in the course work is as follows-
Supervisor’s Sign
(Dr. M.B.Patel)
Grade Obtained in Research Methodology
(PH001)
Grade Obtained in Self Study Course (Core
Subject)
(PH002)
BB BB
vi
Originality Report Certificate
It is certified that PhD Thesis titled “Monitoring and control of Delamination in Drilling
of GFRP (Glass Fibre Reinforced Plastics) by Shri Patel Jaykumar Bipinbhai has been
examined by us. We undertake the following:
a. Thesis has significant new work / knowledge as compared already published or are under
consideration to be published elsewhere. No sentence, equation, diagram, table, paragraph or
section has been copied verbatim from previous work unless it is placed under quotation
marks and duly referenced.
b. The work presented is original and own work of the author (i.e. there is no plagiarism). No
ideas, processes, results or words of others have been presented as Author own work.
c. There is no fabrication of data or results which have been compiled / analysed.
d. There is no falsification by manipulating research materials, equipment or processes, or
changing or omitting data or results such that the research is not accurately represented in the
research record.
e. The thesis has been checked using Turnitin software (copy of originality report attached)
and found within limits (9%) as per GTU Plagiarism Policy and instructions issued from time
to time (i.e. permitted similarity index <=25%).
Signature of the Research Scholar: …………………………… Date: ….………
Name of Research Scholar: Patel Jaykumar Bipinbhai
Place: Ahmedabad
Signature of Supervisor: ……………………………… Date: ………………
Name of Supervisor: Dr. M.B.Patel
Place: Ahmedabad
vii
viii
ix
PhD THESIS Non-Exclusive License to
GUJARAT TECHNOLOGICAL UNIVERSITY
In consideration of being a PhD Research Scholar at GTU and in the interests of the
facilitation of research at GTU and elsewhere, I, Patel Jaykumar Bipinbhai having
Enrolment no. 119997119010 hereby grant a non-exclusive, royalty free and perpetual
license to GTU on the following terms:
a) GTU is permitted to archive, reproduce and distribute my thesis, in whole or in part, and/or
my abstract, in whole or in part (referred to collectively as the “Work”) anywhere in the
world, for non-commercial purposes, in all forms of media;
b) GTU is permitted to authorize, sub-lease, sub-contract or procure any of the acts
mentioned in paragraph (a);
c) GTU is authorized to submit the Work at any National / International Library, under the
authority of their “Thesis Non-Exclusive License”;
d) The Universal Copyright Notice (©) shall appear on all copies made under the authority of
this license;
e) I undertake to submit my thesis, through my University, to any Library and Archives. Any
abstract submitted with the thesis will be considered to form part of the thesis.
f) I represent that my thesis is my original work, does not infringe any rights of others,
including privacy rights, and that I have the right to make the grant conferred by this non-
exclusive license.
g) If third party copyrighted material was included in my thesis for which, under the terms of
the Copyright Act, written permission from the copyright owners is required, I have obtained
such permission from the copyright owners to do the acts mentioned in paragraph (a) above
for the full term of copyright protection.
x
h) I retain copyright ownership and moral rights in my thesis, and may deal with the
copyright in my thesis, in any way consistent with rights granted by me to my University in
this non-exclusive license.
i) I further promise to inform any person to whom I may hereafter assign or license my
copyright in my thesis of the rights granted by me to my University in this non-exclusive
license.
j) I am aware of and agree to accept the conditions and regulations of PhD including all
policy matters related to authorship and plagiarism.
Signature of the Research Scholar:
Name of Research Scholar: Patel Jaykumar Bipinbhai
Date: Place: Ahmedabad
Signature of Supervisor:
Name of Supervisor: Dr.M.B.Patel
Date: Place: Ahmedabad
Seal:
xi
Thesis Approval Form
The viva-voce of the PhD Thesis submitted by Shri Patel Jaykumar Bipinbhai (Enrollment
No. 119997119010 ) entitled “Monitoring and control of Delamination in Drilling of
GFRP (Glass Fibre Reinforced Plastics)” was conducted on …………………….…………
(Day and date) at Gujarat Technological University.
(Please tick any one of the following option)
The performance of the candidate was satisfactory. We recommend that he/she be
awarded the PhD degree.
Any further modifications in research work recommended by the panel after 3 months
from the date of first viva-voce upon request of the Supervisor or request of
Independent Research Scholar after which viva-voce can be re-conducted by the same
panel again.
The performance of the candidate was unsatisfactory. We recommend that he/she
should not be awarded the PhD degree.
---------------------------------------------------- ----------------------------------------------------
Name and Signature of Supervisor with
Seal
1) (External Examiner 1) Name and
Signature
--------------------------------------------------- ----------------------------------------------------
2) (External Examiner 2) Name and
Signature
3) (External Examiner 3) Name and
Signature
xii
Abstract
Composite materials possess several desirable properties when compared against
conventional metal such as, their high specific strength and specific modulus, their variable
directional strength properties and their better fatigue strength. The fiber reinforced plastics
(FRP) are highly promising materials for the applications in aeronautical and aerospace
industries. Composites are being abrasive, the tool wear is high and hence the machining
parameters are to be carefully selected while machining GFRP composite materials.
Machining of these composites, especially drilling is very important operation which
is to be carried out for assembly of composite parts. During drilling of GFRP and CFRP
delamination is a major concern, which reduces the structural integrity of the material. The
present work is critical review which focuses on the analysis of delamination behaviour of the
composites when drilled and methods available to reduce the delamination. Remarkable work
has been carried out by different researchers in this area where few have suggested
controlling the cutting parameters like cutting speed, feed and depth of cut others have
emphasis on thrust force and torque. In this experiment work GFRP laminate with specific
properties is manufactured and drilled with three drills having different point angles of 1180,
1300, 1400. Speed, Feed rate and Point angle are taken as variables and drilled holes are
measured for delamination. Effect of cutting parameters and tool geometry (point angle) on
delamination are studied. ANFIS (Adaptive Neuro Fuzzy Inference System) based
mathematical model is developed in MATLAB to control the delamination. Based on the
above ANFIS model one can select best cutting parameters to minimise delamination.
xiii
Acknowledgement
In pursuit of this academic endeavour, I feel that I have been especially fortunate as
inspiration, guidance, direction, cooperation, love and care all came in my way in abundance
and it seems almost an impossible task for me to acknowledge the same in adequate terms.
It gives me enormous pleasure to express my first thanks and sincere gratitude to my
supervisor Dr. M.B.Patel, for his insistent help and constant guidance; otherwise, it would not
have been possible for me to complete this work. I am obliged to express a deep debt of
gratitude to him. He has helped me from prologue to epilogue. I remain forever grateful to
him.
Besides my supervisor, I would like to express my heartiest thankfulness to the members of
my Doctoral progress Committee (DPC): Prof. Mangal. G. Bhatt, Principal Shantilal Shah
Engineering College, Bhavnagar and Prof. Jitendra A. Vadher, Head of the Department,
Government Engineering College, Palanpur for their kind cooperation and insightful
suggestions throughout period of my project work which has been proved extremely fruitful
for the success of this research work.
I am also highly obliged to Prof. Navneet Khanna, Head of the Department, IITRAM,
Ahmedabad for their academic support and continuous motivation. My special thanks to the
faculty and staff members of Central Workshop of IITRAM, Ahmedabad for their endless
support and continuous assistance.
I would like to thank Mechanical Department, CHARUSET University for extending their
laboratory facilities.
My sincere & deepest gratitude stretches its way to all the friends and motivators for sharing
their valuable information extending their experience of technical expertise and also giving
their valuable time for guiding me, without which this research work would have been
incomplete.
xiv
There goes a popular maxim, “Other things may change us, but we start and end with
family”. Parents are next to God and I would like to thank my parents Mr. B.S.Patel and
Kokila Patel for their blessings and ever increasing unconditional love for me. I would like to
express my sincerest appreciation to my loving wife Nehal for her constant help and
encouragement.
- Jay Patel
xv
Table of Content
Chapter – 1 Introduction 1
1.1 Overview of Composites 1
1.2 Classification of Composites 2
1.3 Fiber Reinforced Composites/Fibre Reinforced polymer
(FRP) Composites 3
1.3.1 Reinforcements 3
1.3.2 Matrix 3
1.3.3 Advantages of Composites 4
1.3.4 Limitations of Composites 4
1.4 Fabrication Processes of Fibre Reinforced Polymers
Composites 4
1.4.1 Hand Lay-Up Process 5
1.4.2 Spray Lay-Up Process 5
1.4.3 Filament Winding Process 6
1.4.4 Vacuum Infusion Process 7
1.4.5 Autoclave Moulding 7
1.5 Properties of GFRP Composites 8
1.6 Applications 9
1.7 Machining of Composites 10
1.7.1 Types of Machining in GFRP 11
1.7.2 Defects and Problems Encountered in Drilling of
Composites 12
1.8 Research Overview 15
1.9 Organisation of Thesis 16
Chapter – 2 Literature Review 18
2.1 Machining of GFRP 18
2.2 Drilling of GFRP 20
2.3 Drilling parameters 22
2.3.1 Effect of tool materials, types and its geometry
on thrust force and torque 23
2.3.2 Effect of tool materials and its geometry on
xvi
delamination 24
2.3.3 Effect of cutting parameters (speed and feed
rate) on thrust force and torque 25
2.3.4 Effect of cutting parameters (speed and feed
rate) on delamination 27
2.3.5 Effect of combination of tool and process
Parameters 30
2.4 Modelling of Drilling Parameters 31
2.4.1 Response Surface Methodology 31
2.4.2 Fuzzy Logic 32
2.4.3 ANFIS (Adaptive Neuro Fuzzy Inference System) 33
2.5 Optimization 35
2.5.1 Grey relational analysis 35
2.5.2 Grey fuzzy approach 37
2.6 Motivation for the Research 38
2.7 Problem Formulation 39
2.8 Scope and Objectives 40
2.8.1 Objectives 40
2.8.2 Methodology 40
Chapter – 3 Experimental Work 42
3.1 Specimen Fabrication and Material Properties 42
3.1.1 Fabrication 43
3.1.2 Testing 44
3.2 Selection of Tool and Tool Geometry 50
3.3 Selection of Drilling Parameters 51
3.3.1 Spindle Speed 52
3.3.2 Feed Rate 53
3.3.3 Point Angle 53
3.4 Design of Experiment 53
3.5 Experimental Set Up 54
3.6 Measurement of Responses 63
3.6.1 Thrust force 63
3.6.2 Measurement of Torque 65
xvii
3.6.3 Calculation of Delamination factor 65
3.7 Experimental Observations 68
3.8 Summery 70
Chapter – 4 Results and Discussion 71
4.1 Adaptive Neuro Fuzzy Inference System (ANFIS) 71
4.2 Modelling Drilling Parameters Using ANFIS 75
4.3 Effect of Input Variables in Drilling of GFRP 77
4.3.1 Effect of Drilling Parameters on Thrust Force 78
4.3.2 Effect of Drilling Parameters on Torque 80
4.4 ANOVA Analysis of Drilling Parameters and Plots
Showing the Interaction Effect of Drilling Parameters in
Drilling of GFRP Composites 81
4.4.1 Analysis of Thrust Force using ANOVA 81
4.4.2 Analysis of Delamination at Entry (Fdi) using
ANOVA 84
4.4.3 Analysis of Delamination at Exit (Fdo) using
ANOVA 87
4.5 Summery 90
Chapter – 5 Conclusions and Future Scope 91
5.1 Conclusions 91
5.2 Future scope 93
References 94
List of Publications 101
xviii
List of Abbreviation
GFRP - Glass Fibre Reinforced Plastics
ANFIS - Adaptive Neuro Fuzzy Inference System
MMC - Metal Matrix Composites
OMC - Organic Matrix Composites
CMC - Ceramic Matrix Composites
FRP - Fibre Reinforced composites
PMC - Polymer Matrix Composites
CFRP - Carbon Fibre Reinforced Plastics
AFRP - Aramid Fibre Reinforced Plastics
UD - Unidirectional
PCD - Poly-Crystalline Diamond
CBN - Cubic Boron Nitride
DOE - Design of Experiment
ANOVA - Analysis of Variance
ANN - Artificial Neural Network
LFVAD - Low Frequency Vibration Assisted Drilling
ANOM - Analysis of Means
SEM - Scanning Electron Microscopy
RSM - Response Surface Methodology
FIS - Fuzzy Inference System
PSD - Power Spectral Density
MRR - Material Removal Rate
ATIRA - Ahmedabad Textile and Industrial Research Association
HSS - High Speed Steel
DoF - Degrees Of Freedom
DAQ - Data Acquisition System
MF - Membership Function
GUI - Graphic User interface
TD - Twist Drill
SD - Step Drill
MFD - Multifaceted Drill
xix
List of Symbols
Fda - Adjusted delamination factor
Fz - Thrust force in N
Fx - Force in X direction in N
Fy - Force in Y direction in N
MZ - Torque in N.mm
V - Spindle speed in rpm
f - Feed Rate in mm/min
θ - Point Angle in degrees
L - Length in mm
W - Width in mm
H - Height in mm
Fd - Delamination Factor
Fdi - Delamination at Entry
Fdo - Delamination at Exit
Dmax - Maximum diameter of the hole in mm
D - Drill diameter in mm
Amax - Maximum Area of the hole considering damage in mm2
A - Area of actual hole drilled in mm2
Ai - linguistic label associated with node function
xx
List of Figures
Sr no. Title Page no.
1.1 Hand lay-up process 5
1.2 Spray lay up process 6
1.3 Filament Winding Process 6
1.4 Vacuum Infusion Process 7
1.5 Autoclave Moulding 8
1.6 Aircraft parts made from composites 10
1.7 Delamination phenomenon 13
1.8 Development of spalling effect 14
2.1 Methodology 41
3.1 Vacuum Infusion Process 43
3.2 (a) Standard specimen for tensile test- Actual specimen 44
3.2 (b) Standard specimen for tensile test- Specimen geometry 44
3.3 Specimen Testing for Tensile strength 46
3.4 (a) Tensile test report 47
3.4 (b) Compressive test report 48
3.4 (c) BARCOL Hardness test report 49
3.5 (a) Twist drill 51
3.5 (b) Step drill 52
3.5 (c) Multifaceted Drill (8 facets) 52
3.6 Schematic diagram of experiment set up 56
3.7 Actual experiment set up 56
3.8 Fixture to hold the laminate 60
3.9 Dynamometer fitted under fixture 60
3.10 Charge amplifier and DAQ 61
3.11 Actual drilling 61
3.12 Actual drilling without coding 62
3.13 Actual drilling with coding 62
3.14 Force FX versus Time 63
3.15 Force FY versus Time 64
3.16 Force FZ versus Time 64
3.17 Torque MZ versus Time 64
3.18 Measurement of delamination using IMAGE J software 66
3.19 Measurement of drilled hole on 3D Microscope 67
xxi
List of Figures
Sr
no. Title Page no.
4.1 Sugeno Fuzzy inference model with three inputs (x, y, z) and one
output (f) 72
4.2 ANFIS Architecture for three inputs (x, y, z) and one output (f) 73
4.3 Structure of Sugeno type FIS model with three inputs and one output 76
4.4 ANFIS Structure for three input and one output 76
4.5 Rules for three input and one output 78
4.6 Thrust force obtained at 100 m/min feed rate and speed of 1500, 2000
and 2500 rpm 79
4.7 Torque at 100 m/min feed rate and speed of 1500, 2000 and 2500 rpm 80
4.8 Effect of spindle speed, point angle and feed rate on thrust force 83
4.9 Effect of spindle speed, point angle and feed rate on Delamination at
Entry 86
4.10 Effect of spindle speed, point angle and feed rate on Delamination at
Exit 88
xxii
List of Tables
Sr no. Title Page no.
1.1 Properties of E-glass and S-glass 8
1.2 Mechanical properties of GFRP, CFRP and AFRP 9
3.1 Mechanical properties of GFRP laminates 45
3.2 Tool and its geometry details 51
3.3 Process parameters and their levels for experiments 54
3.4 Taguchi’s L27 Orthogonal Array 55
3.5 Specification of CNC vertical machining centre (Macpower) 57
3.6 Specification of Kistler Dynamometer Type 9272 58
3.7 Specification of Multichannel charge Amplifier Type 5070A 59
3.8 Specification of Data Acquisition System for Force Measurement 59
3.9 Specification of 3D microscope 67
3.10 Observation table for thrust force and delamination at entry Fdentry 68
3.11 Observation table for thrust force and delamination at exit Fdexit 69
4.1 ANFIS information for different MFs 77
4.2 ANOVA table for Thrust force F 81
4.3 ANOVA table for Delamination at Entry Fdi 85
4.4 ANOVA table for Delamination Factor at Exit Fdo 87
4.5 Comparison of predicted and experimented values 90
Introduction
1
CHAPTER – 1
Introduction
Present chapter provides overview to composites, Glass Fibre Reinforced Plastics (GFRP),
Machining of GFRP in context with drilling, defects generates in drilling of GFRP,
Different parameters affecting the drilling of GFRP, objectives of the research, research
methodology, and research overview.
1.1 Overview of Composites
Fibres or particles embedded in matrix of another material are known as composites.
Laminates are composite material where different layers of materials give them the specific
character of a composite material to perform a specific function. Fabrics have no matrix
but in them, fibres of different compositions combine to give them a specific character.
Reinforcing materials generally withstand maximum load and serve the desirable
properties. Further, though composite types are often distinguishable from one another, no
clear determination can be really made. To facilitate definition, the accent is often shifted
to the levels at which differentiation take place viz., microscopic or macroscopic.
In matrix-based structural composites, the matrix serves two paramount purposes viz.,
binding the reinforcement phases in place and deforming to distribute the stresses among
the constituent reinforcement materials under an applied force.
The demands on matrices are many. They may need to temperature variations, be
conductors or resistors of electricity, have moisture sensitivity etc. This may offer weight
advantages, ease of handling and other merits which may also become applicable
depending on the purpose for which matrices are chosen.
Introduction
2
1.2 Classification of Composites
Composite materials are commonly classified at following two distinct levels:
• The first level of classification is usually made with respect to the matrix constituent.
The major composite classes include Organic Matrix Composites (OMCs), Metal Matrix
Composites (MMCs) and Ceramic Matrix Composites (CMCs). The term organic matrix
composite is generally assumed to include two classes of composites, namely Polymer
Matrix Composites (PMCs) and carbon matrix composites commonly referred to as
carbon- carbon composites.
• The second level of classification refers to the reinforcement form - fibre reinforced
composites, laminar composites and particulate composites. Fibre Reinforced composites
(FRP) can be further divided into those containing discontinuous or continuous fibres.
• Fibre Reinforced Composites are composed of fibres embedded in matrix material. Such
a composite is considered to be a discontinuous fibre or short fibre composite if its
properties vary with fibre length. On the other hand, when the length of the fibre is such
that any further increase in length does not further increase, the elastic modulus of the
composite, the composite is considered to be continuous fibre reinforced. Fibres are small
in diameter and when pushed axially, they bend easily although they have very good
tensile properties. These fibres must be supported to keep individual fibres from bending
and buckling.
• Laminar Composites are composed of layers of materials held together by matrix.
Sandwich structures fall under this category.
• Particulate Composites are composed of particles distributed or embedded in a matrix
body. The particles may be flakes or in powder form. Concrete and wood particle boards
are examples of this category.
Fibres are the important class of reinforcements, as they satisfy the desired conditions and
transfer strength to the matrix constituent influencing and enhancing their properties as
desired.
Glass fibres are the earliest known fibres used to reinforce materials. Ceramic and metal
fibres were subsequently found out and put to extensive use, to render composites stiffer
more resistant to heat.
Introduction
3
1.3 Fiber Reinforced Composites/Fibre Reinforced Polymer (FRP)
Composites
1.3.1 Reinforcements
A fibrous material which is nonwoven, strong and incorporated in the matrix material to
improve its physical properties is called reinforcements. Carbon, Glass, Graphite,
Asbestos, Boron, Jute, Chopped paper, Synthetic fibre etc. are used as reinforcements.
Reinforcements are used to strengthen tensile and flexural strength while filler have no
specific purpose. Reinforcement creates a bond to make the joint stronger. Main purpose of
reinforcement is to improve the mechanical properties of the resin system. All above said
reinforcement’s fibres have different properties and affects the properties of composite in
different ways. Reinforcements provide strength to composite and also used to increase the
heat resistance, corrosion resistance and rigidity.
Selection of fibres depends upon certain important points which are compatibility with
matrix, density of overall composite, melting point temperature thermal stability and so on.
1.3.2 Matrix
There is no doubt about it that the original high strength of composites is due to its fibres
but the role of matrix is also equally important in imparting strength to composites because
it supports the fibres and helps in carrying the loads.it also acts as binding agent and
provides stability to the composite materials. If more resin is used than fibres than the
composite is known as resin rich and if it is less than fibres it is known as fibre rich or resin
starved composite. When the resin is more it is more susceptible to cracking and resin
starved composite has less strength as fibres will not be arranged properly and will not
have enough support of resin. Few functions of the fibres are it holds the fibres together,
environment protection, distribution of loads in fibres, fracture resistance and impact
resistance is improved, avoids propagation of cracks, to enhance transverse properties of
composites. Desired properties of matrix are it should reduce moisture absorption, low
coefficient of thermal expansion, it should be elastic to transfer loads, chemical resistance,
good strength, modulus and elongation. Epoxy, phenolic, polyester, polyurethane, vinyl
ester are examples of resin.
Introduction
4
1.3.3 Advantages of Composites
Composites are widely used in aerospace industry due to its following advantages:
Stiffness to weight ratio is very high
Higher reliability so lesser inspections and repairs
Resistance to corrosion and fatigue is high.
It can be designed according to the requirement of strength by putting the fibres in
certain patterns to withstand the applied loads.
Smooth aerodynamic properties can be achieved
Good torsional stiffness can be achieved
Resistance to impact damage is high
Good dimensional stability
Good weather ability
1.3.4 Limitations of Composites
Some of the disadvantages of composites are:
Fabrication and Raw material cost is high
Composites have more brittleness
Toughness is less due to matrix weakness.
Composites are difficult to dispose and reuse
Repairing is not possible
Difficult to analyse
Environment degradation is possible
1.4 Fabrication Processes of Fibre Reinforced Polymers Composites
There are many manufacturing processes available for making composites of required
strength and desired performance in an economic way. Different manufacturing processes
available are hand lay-up process, spray up process, vacuum pressure process, filament
winding, continuous pultrusion, injection moulding, resin transfer moulding, matched die
moulding etc.
Introduction
5
1.4.1 Hand Lay-Up Process
It is the oldest and most common method of fabricating composite structural laminates.
The mould is used to give the shape to structures and fibres and resins are placed on it
layer by layer to give it thickness and roller is used to remove air from the material. In the
hand lay-up curing is done at room temperature. A gel coat is applied to achieve very high
quality. Generally a surface which is exposed to the air is rough but can be taken care of by
suitable wiping films. Epoxies and polyesters are mainly used as resin in hand lay-up
process. Figure 1.1 shows the Line diagram of hand lay-up process.
FIGURE 1.1 Hand lay-up process
1.4.2 Spray Lay-Up Process
This process is an extension of the hand lay-up process. Specialised spray gun is utilised in
this technique to spray the resin under pressure and for reinforcement which is in the form
of chopped fibres. As shown in figure 1.2 the resin and fibre can be sprayed
simultaneously or it can be added separately. A glass roving is suppled to spray gun where
it is chopped by chopper gun. Here also a roller is used to remove air from the material.
After spraying number of layers for required thickness curing is done at room temperature.
Any size part can be made by using this technique with good volume fraction
reinforcement.
Introduction
6
FIGURE 1.2 Spray lay up process
1.4.3 Filament Winding Process
To utilise the fibre strength effectively continuous reinforcement is required which is done
by filament winding process.a creel is used to feed the fibre through the reservoir of resin
and wound on the other side on mandrel. To get the maximum strength in specific direction
fibres are lay down in a pattern by using special winding machines. The composites made
by filament winding is cured at room temprature after required layers have been layed.
Figure 1.3 shows the schematic diagram of Filament winding process.
FIGURE 1.3 Filament Winding Process
Introduction
7
1.4.4 Vacuum Infusion Process
Composite material of highest quality with consistantncy can achieved in vacuum infusion
process.it produces very less amount of hazardous air pollutants. Vacuum infusion process
uses vacuum pressure to drive resin into a laminate. As a reinforcement any fibre can be
used and a flexible bag is used to coer the mould. To draw the resin in a specific area in the
mould the boundries of the mould are sealed and vacuum is used to give direction to resin
from its container.the advantages of using vacuum infusion process are void content is less
and better fiber to resin ratio is achieved accurately, Less amount of hazardous air
pollutants Good health and safety and Very consistent resin usage. Figure 1.4 shows the
vacuum infusion process.
FIGURE 1.4 Vacuum Infusion Process
1.4.4 Autoclave Moulding
Controlled pressure and heat is given in the technic of Autoclave moulding. Higher density
and removal of volatiles from the resin in curing is ensured by application of pressure.
Both side of the mould can be smooth because one side mouldis of steel while other side
mould is of nylon film. Reinforcing fibres can be placed enywhere in the mould manually
based on required strength. The process is performed at higher temparature and pressure
then atmospheric so that higher volume fraction ratio can be achieved. Maximum structural
efficiency and minimum reaction with atmosphere can be achieved in this process. Figure
1.5 shows the diagram of Autoclave moulding.
Introduction
8
FIGURE 1.5 Autoclave Moulding
1.5 Properties of GFRP Composites
Glass fibre reinforced plastics has a unique properise. It has a wide range of apllications.its
dimentional and functional propertise make it useful for various engineering
applications.E-glass is the most commonly used glass fibre. It is a lime,aluminium and
borosilicate glass with minimum sodium and potasium levels. S-glass is having higher
strength to weight ratio and is more costly than E-glass is mostly used for aerospace and
military applications.
TABLE 1.1 Properties of E-glass and S-glass
Composites are having better properties in comparison to conventional materials i.e. metals
like high specific strength and high modulus, variable directional strength and fatigue
E-glass S-glass
Composition 52 to 56 percent SiO2
12 to 16 percent Al2o3
16 to 25 percent CaO
8 to 13 percent B2O3
65 percent SiO2
25 percent Al2o3
10 PERCENT MgO
Tensile strength 500 ksi (3.44 GPa) 350 ksi (4.48 GPa)
Modulus of elasticity 10.5 Msi (72.3 GPa) 12.4 Msi (85.4 GPa)
Introduction
9
strength. These properties can be achieved by proper mix of fibres and resin. Table 1.1
shows the properties of different FRPs.
Table 1.2 Mechanical properties of GFRP, CFRP and AFRP (Source: R.Teti,
‘Machining of composite materials’. CIRP)
FRP material Tensile
strength(MPa)
Elastic
modulus
MPa)
Strain to
failure
Density
(g/cm3)
GFRP
Unidirectional
Vf=60% 1000 45,000 2.3 2.1
Vf=20%-50%
Woven cloth 100-300 10000-20000 - 1.5-2.1
CFRP
Unidirectional
Vf=60% High
strength
1200 145000 0.9 1.6
Unidirectional
Vf=60% High
modulus
800 220000 0.3 1.6
AFRP Unidirectional
Vf=60% 1000 75000 1.6 1.4
1.6 Applications
GFRP is a most common type of composite and is first used in 1950 for boats and
automobiles and today it is widely used in modern cars. Glass fibre reinforced plastics was
first used in Boeing 707 in 1950 and it was comprising of 2.5 to 3 percent of total volume
of structure. From such a beginning, composite applications has made a revolution in
industries like aerospace, marine and electrical, chemical and transportation etc. The
composite industry have recognise now that composites have extraordinary potential for
business opportunity in any applications. Figure 1.6 shows the aircraft parts which are
made from different type of composites.
Not only in aircraft industry but in machine tools, sports industry and in automobiles
composites have replaced conventional materials. Higher mileage and efficiency and lower
fuel consumption requirement can only be achieved by using composites in automobiles.
Heat absorption and dissipation problems in printed circuit boards are solved using
composites in computer industry. Prosthetic limbs, joint replacements and certain other
implants have become possible using composites. Bullet proof jackets and helmets from
Introduction
10
carbon fibres and aramid fibres are the requirement of today’s militaries. In sports industry
mountain bikes frames are made from carbon fibre reinforced plastics, tennis rackets,
trekking and hiking equipment have made those sports more competitive. Corrosion
resistance and stiffness properties have made composites as an ideal choice in structural
engineering. Fishing rods are manufactured from composites. Boats weights are reduced
due to composites. In telecommunication industry transmitting towers of metals are also
being replaced with composites. In wind turbines and towers composites are inevitable.
(Source: http://www.aml.engineering.columbia.edu/ntm/level1/ch05/html/l1c05s03.html)
FIGURE 1.6 Aircraft parts made from composites
1.7 Machining of Composites
Composites required different types Machining operations to get them in required shape
and size though it is fabricated near to required shape. To bring the composites in required
geometrical and dimensional tolerances machining is inevitable. Machining of composites
is different from that of conventional machining in a way that composites are highly
abrasive and having higher tool wear compared to conventional machining. So that the
cutting parameters to be selected for machining is having highest importance. Wang and
Zhang et al (2003) performed an experimental investigation into the orthogonal cutting of
Introduction
11
UD FRP [1]. The outcome was that the fiber orientation is responsible for the damage on
the surface and its mechanisms in a machined component. As the fibre orientation changes
the cutting forces also changes with surface roughness and subsurface damage. It was
studied machinability do not have any effect of the cutting conditions of preparing
composites.
1.7.1 Types of Machining in GFRP
Different conventional and non-conventional machining operations are essential for
bringing the composites in final shape and size. Conventional machining are turning,
drilling, grinding, milling, shaping etc. while non-conventional machining are water jet
machining, laser beam machining and electrical discharge machining.
Turning:
Davim et al (2001) investigated the effect of cutting parameters on the surface finish in
turning. He concluded that the cutting velocity has greater influence on the surface
roughness followed by feed and there is no significant influence with depth of cut.
However, the interaction of velocity and feed is the most important of other analysed
parameters [2].
Hussain, Pandurangadu, and Palani Kumar et al (2011) studied of GFRP composite’s
machinability for fiber orientation from 300 to 900. All geared lathe is used for machining
purpose with three different cutting tools which are Poly-Crystalline Diamond (PCD),
Carbide (K-20) and Cubic Boron Nitride (CBN). Taguchi’s Design of Experiments (DOE)
L25 orthogonal array is employed for conducting the on an all geared lathe. Cutting speed,
depth of cut, feed rate and work piece (fiber orientation) were taken as cutting parameters.
Surface roughness (Ra) and Cutting force (Fz) were measured to evaluate the performances
of the cutting tools. Response Surface Methodology is used to develop a second order
mathematical model in terms of cutting parameters. The developed model can be applied to
predict the cutting forces and surface roughness in machining of Glass Fibre Reinforced
composites [3].
Milling:
Hussein, Asif and Li investigated the influence of drilling and milling parameters in drill
making on glass fibre reinforced laminates. ANOVA is used to understand separately the
Introduction
12
effects of cutting parameters in drilling and milling. It is found that when cutting quality is
of importance milling operation is more suitable in than drilling at high cutting speed and
low feed rate [4].
Grinding:
Hu and Zhang et al (2004) studied the grinding operation of UD CFRP composite materials
using an alumina grinding wheel. It was investigated when fibres are oriented at 60° and
90° grinding forces are higher, but when fibres are oriented at 120° and 180° grinding
surfaces are poor. Fiber orientation, depth of grinding are responsible for surface integrity,
which is equal to the findings of orthogonal cutting [5].
Drilling:
Lachaud and Francis (2001) proposed a model which links the axial penetration of the drill
bit to the conditions of delamination at exit (last few plies). Many types of tool/material
contact conditions were studied compared with experimental measurements. They have
established a close correlation between experiment and calculation when the thrust force of
the drill is modelled by taking into account the geometrical nature of the contact between
the tool and a laminate composite material [6].
1.7.2 Defects and Problems Encountered in Drilling of Composites
Due to the anisotropic property, plastic deformation characteristic and abrasive property
drilling of GFRP composites are more challenging. Many researchers have carried out
experiments and faced following problems while drilling the composites:
Delamination:
In glass fibre reinforced plastics delamination is a major failure mechanism which is one of
the important factor to differentiate it from metals. High inter laminar stresses and less
through thickness strength is a major cause for delamination. This phenomenon occurs due
to fibres are lying in the same plane of the laminate and so do not provide enough
reinforcement. Comparatively weak matrix has to carry all loads instead of fibres.
Delamination failure can be judge by the sound produced. Delamination can be defined as
the ratio of maximum diameter of the hole including defects to the actual diameter of the
hole or drill.
Introduction
13
maxd
nom
DF
D
(1.1)
Where Fd = delamination factor
Dmax= Maximum diameter of the hole
D = Drill diameter
Delamination can be at entry and at exit of the laminate. Entry delamination is known as
peel up delamination while exit delamination is known as push out delamination which is
shown in figure 1.7. Hocheng and Dharan (1990) investigated that damage takes place both
at the entry and the exit of the drill and thus differentiated the damage as peel-up at
entrance and push-out at the exit [7].
(a) Delamination at entry
(b) Delamination at exit
FIGURE 1.7 Delamination phenomenon
Spalling and fuzzing:
Spalling is related to the delamination defect and fuzzing is related to fibres which ae not
cut properly. Zhang et al (2001) developed an empirical relationship between cutting
parameters and area of delamination and described the fuzzing damage in numerical value.
Introduction
14
Figure shows the schematic diagram of development of spalling effect [8].
FIGURE 1.8 Development of spalling effect
Spalling and Fuzzing have linear relationship, as spalling increases the fuzzing effect also
increases.
Matrix burning, de-bonding and fibre pull out:
Plate bulge, crack opening, fibre twisting and fibre tearing are three different mechanism
identified by Dipaolo et al (1996) [9]. Matrix burning, de-bonding and fibre pull out are
investigated as major sources of damage by Mathew et al (1999) [10]. Piquet et al (2000)
have carried out series of experiments and found that a conventional drill a tool made from
micro grain tungsten carbide having smaller rake angle can reduce the defects like
delamination at entry, delamination at exit, fibre bending and buckling, shear failure of
fibre and error of roundness of hole.it was investigated that roundness error is due to the
anisotropy of material [11]. Chen et al (1997) suggested that correct selection of tool
geometry and cutting parameters can lead to delamination free drilling [12].
Hocheng and Tsao et al (2006) have studied non-conventional methods, special types of
drills, pilot hole making, using back up plate and step drilling to reduce the different kind
of defects and damages produced during drilling [13].
Zitoune and Collombet et al (2007) have applied finite element analysis to find the
responsible thrust force at the exit of the hole while drilling long fibred composites. The
tool geometry and effect of shear force were considered in the developed finite numerical
model with compare to other analytical models. The numerical model was validated by
drilling on long fibred carbon epoxy laminates. Numerical outcome had correlation with
the experimental results [14].
Introduction
15
Adjusted delamination factor (Fda) was measured using a novel technique by Paulo
Davim, Rubio and Abrao et al (2007). Experimental design is also introduced for drilling
GFRP sheets with specific drilling parameters. Adjusted delamination factor (Fda) was
calculated by digital analysis. The digital analysis and experimental results have proved
that it is a good technique to estimate adjusted delamination factor (Fda) [15].
Palanikumar, Rubio, Abrao and Davim (2008) have proposed a mathematical model to
predict delamination in drilling glass fiber-reinforced plastic composites [16].
1.8 Research Overview
Glass Fibre Reinforced Plastics (GFRP) are having the properties of light weight, higher
strength to weight ratio, good fatigue resistance and high modulus. According to the
specific application these properties of GFRP can be fabricated. GFRP are used in
aerospace, automobiles, sports goods, satellite and military equipment, telecommunication
and marine industries. For the particular application, composite structures are moulded to
near-net shape but for final assembly, surface finish and dimensional accuracy machining
operations are required to perform. Different machining operations are carried out for
different purposes but drilling is an essential operation to be carried out to assemble the
parts by riveting, screwing or bolting. In this research work drilling operation is carried out
on Glass Fibre Reinforced Plastics (GFRP) laminates of 4 mm average thickness.
Three types of drills having tool geometry of standard Twist drill, Step drill and
multifaceted drill are manufactured with point angle of 1400, 1300 and 1180 respectively
for experimentation work. The experiment is designed using Taguchi’s L27 orthogonal
array of experiments using MINITAB software. The cutting parameters selected are
spindle speed, feed rate and point angle and the experiment was carried out on Macpower
make computer numerical control machining centre. Piezoelectric dynamometer, Amplifier
and Data acquisition system of KISTLER make is used to measure thrust force and torque.
IITRAM laboratory facilities were used to carry out at the experiment. The laminate plate
is drilled for 81 holes and each hole is observed under the 3 D microscope of Mitutoyo
make at CHAURSET University. Delamination factor at entry and exit of the hole are
calculated based on the observed readings. The effect of cutting parameters i.e. spindle
speed, feed rate and point angle were studied on thrust force, torque and delamination
Introduction
16
factor. To decide the significance of factors affecting the process and their inter relation
which affect the process, ANOVA analysis is performed. A mathematical model using
ANFIS (Adaptive Neuro Fuzzy Inference System) is developed using MATLAB software.
The model has the advantage of fuzzy logic as well as ANN (Artificial Neural Network).
1.9 Organisation of Thesis
Chapter 1 includes an overview of composite materials, classification of composites,
constituents of composites, fabrication methods available and its applications. Importance
of glass fibre reinforced plastics is discussed. Though composites are moulded near to final
shape, machining of composites is essential hence machining operations like turning,
drilling, milling, grinding etc. are discussed and damages and defects encountered while
machining composites are explained in detail. Main defect is delamination which has to be
minimised in drilling of GFRP is the focus of this research so factors affecting
delamination are identified. The overview of the research is presented.
Chapter 2 presents literature review studied on glass fibre reinforced plastics, machining of
GFRP, drilling of GFRP and the parameters affecting the process. Delamination is studied
from the view of tools, tool material and its geometry. Responsible factors for
delamination i.e. thrust force and feed rate and related articles were referred and their
effects on delamination were studied in this chapter. Moreover the experiment work and
theories developed in the field of GFRP machining are understood thoroughly and their
results are discussed. Optimisation theories, ANOVA (Analysis of variance) and ANFIS
(Adaptive Neuro Fuzzy Inference System) are explored. Thus the detailed study is carried
out in the field of machining of composites and based on that the problem is identified,
problem definition is formulated and scope and objectives are fixed. Finally the
methodology of ongoing research is presented.
Chapter 3 represents the details of actual experimentation work carried out. Fabrication of
GFRP laminates and finalising its properties, testing of composite plates for the required
properties, its reports and the specification of testing machines are mentioned in this
chapter. Selection of drill tools and its geometry and fabrication of tools is discussed.
Selection of drilling parameters and their levels for experiment is fixed. Design of
experiment using Taghuchi L27 orthogonal array is performed in MINITAB software.
Introduction
17
Schematic experimental set up is discussed with each equipment in detail and actual
photographs of experiment work are presented. Observation table covering drilling
parameters and the responses obtained in the experiment are listed along with the sift wares
and equipment used to measure the responses.
Chapter 4 describes the results and models developed based on the findings. A model is
developed using soft computing approach of ANFIS (Adaptive Neuro Fuzzy Inference
System) in MATLAB software. ANOVA analysis is performed to decide the significant
factors and their interrelations. Surface plots are created in ANFIS and comparison is done
between the values obtained by experimental work and predicted values using
mathematical model. Experiments are done for validation of the predicted values which are
found inline.
Chapter 5 presents the conclusion of the research work and future scope of the work.
Literature review
18
CHAPTER – 2
Literature Review
Composite materials are the new materials for emerging applications in various
engineering field, because of its mechanical, structural and functional properties. Though
the composite structure are fabricated near to its required shape during the fabrication, the
final dimension and size of the composite structure is achieved by performing machining
operations. One of the major operation required is drilling for the purpose of joining two
different parts of composite structures. Drilling of composites is a bit difficult process due
to its anisotropic properties. To minimize the defects due to drilling is the main objective
of this research work in GFRP composite laminates using optimization of cutting
parameters and tool geometry. Statistical computational techniques are used to the study
and analysis of damage like ANOVA and soft computing approach, ANFIS (Adaptive
Neuro Fuzzy Inference System) to select the cutting parameters for delamination free
drilling. The huge research work carried out in GFRP machining. A literature review of the
Glass fibre reinforced plastics machining, GFRP composite drilling, and analysis of cutting
parameters, optimization techniques and soft computing are discussed briefly in this
chapter.
2.1 Machining of GFRP
Machining operations like turning, milling, drilling and grinding are performed on Glass
Fibre Reinforced Plastics (GFRP) to give its final shape and for assembly purpose.
Davim et al (2001) investigated the effect of cutting parameters on the surface finish in
turning. He concluded that the cutting velocity has greater influence on the surface
roughness followed by feed and there is no significant influence with depth of cut.
However, the interaction of velocity/feed is the most important of other analysed
parameters. Prediction of surface roughness by means of multiple regression analysis
showed lower associated error than that predicted by theoretical model [2].
Literature review
19
Cenna and Mathew (2002) developed a theoretical model that predicted the different
parameters in laser cutting of GFRP composite materials like kerf width at the entry and
exit, Material removal rate and transmission of energy through the cut kerf. However, the
model had underestimated the results obtained for the GFRP composite material; since it
involves a different material removal mechanism than other FRPs [17].
Liakus et al (2003) described simulations of composites produced by a fiber or tow spray
deposition process. A link between composite manufacturing processes and reinforcement
orientation distribution and finally to property predictions was established [18].
Wang and Zhang et al (2003) performed an experimental investigation into the orthogonal
cutting of UD FRP. The outcome was that the fiber orientation is responsible for the
damage on the surface and its mechanisms in a machined component. As the fibre
orientation changes the cutting forces also changes with surface roughness and subsurface
damage. It was studied machinability do not have any effect of the cutting conditions of
preparing composites [1].
Gordon and Hillery (2003) presented a review of the cutting of FRP composite materials.
They identified that most of the research published is concentrated on the chip formation
process and cutting force prediction with unidirectional FRP materials. They identified that
the metal cutting tools and techniques are still used for the most part in the cutting of
composites [19].
Hu and Zhang et al (2004) studied the grinding operation of UD CFRP composite materials
using an alumina grinding wheel. It was investigated when fibres are oriented at 60° and
90° grinding forces are higher, but when fibres are oriented at 120° and 180° grinding
surfaces are poor. Fiber orientation, depth of grinding are responsible for surface integrity,
which is equal to the findings of orthogonal cutting [5].
Davim and Reis (2005) performed ANOVA to investigate the cutting characteristics,
velocity and feed rate, under the surface roughness and damage in milling laminate plates
of CFRP composites. A multivariable regression analysis showed that feed rate has the
highest statistical and physical influence on surface roughness and on delamination factor
[20].
Literature review
20
Hussain, Pandurangadu, and Palani Kumar et al (2011) studied of GFRP composite’s
machinability for fiber orientation from 300 to 900. All geared lathe is used for machining
purpose with three different cutting tools which are Poly-Crystalline Diamond, Carbide
and Cubic Boron Nitride. Taguchi’s Design of Experiments L25 orthogonal array is
employed for conducting the on an all geared lathe. Cutting speed, depth of cut, feed rate
and work piece (fiber orientation) were taken as cutting parameters. Surface roughness
(Ra) and Cutting force (Fz) were measured to evaluate the performances of the cutting
tools. Response Surface Methodology is used to develop a second order mathematical
model in terms of cutting parameters. The developed model can be applied to predict the
cutting forces and surface roughness in machining of Glass Fibre Reinforced composites
[3].
2.2 Drilling of GFRP
Lachaud and Francis (2001) proposed a model which links the axial penetration of the drill
bit to the conditions of delamination at exit (last few plies). Many types of tool/material
contact conditions were studied compared with experimental measurements. They have
established a close correlation between experiment and calculation when the thrust force of
the drill is modelled by taking into account the geometrical nature of the contact between
the tool and a laminate composite material [6].
Davim and Reis (2003) investigated delamination in CFRP composite laminate. HSS and
cemented carbide drills were used and experiments were performed according to Taghuchi
design of experiments. Multiple linear regression technique is used and a direct relation
was found between feed rate and cutting velocity with the delamination. The confirmation
results suggested that the deviation related with the delamination factor has accurate
correlation [21].
Zhang, Lijiang and Xin (2003) suggested a model which can predict critical thrust force
and torque in vibration drilling of fibre reinforced laminates. Vibration drilling method and
hybrid variation parameters method was used to predict the model. High efficiency and
precise quality holes were achieved by vibration drilling [22].
Literature review
21
Edoardo Capello (2004) analysed the differences in delamination mechanisms when
drilling with and without a support placed under the workpiece. The investigation has led
to hypothesize two main differences in the mechanism. Prototype mechanism was
developed based on hypothesized delamination mechanism and it was verified results show
that the Prototype mechanism can drastically lower delamination [23].
Khashaba (2004) investigated experimentally the influence of drilling and material
variables on thrust force, torque and delamination of GFRP composites. Drilling variables
were cutting speed and feed. Material variable were matrix type, filler and fiber shape.
Drilling process was carried out on different materials like cross-winding/polyester,
continuous-winding with filler/polyester, chopped/polyester, woven/polyester and
woven/epoxy composites. Accurate technique was developed to measure delamination size
[24].
Mohan, Ramachandra and Kulkarni (2005) have been conducted Drilling tests on GFRP on
CNC milling centre. Feed rate, and cutting speed and drill size were taken as input
parameters and delamination was measured. After number of experiments on glass fiber-
reinforced polyester laminates thrust force and torque are recorded as responses. Semi
empirical relationship model was developed in terms of cutting parameters. 6 mm drill size
has better correlation than 10 mm diameter drill in model with experimented values. Lower
feed ranges and torque have better correlation than for the higher feed ranges [25].
Zitoune and Collombet et al (2007) have applied finite element analysis to find the
responsible thrust force at the exit of the hole while drilling long fibred composites. The
tool geometry and effect of shear force were considered in the developed finite numerical
model with compare to other analytical models. The numerical model was validated by
drilling on long fibred carbon epoxy laminates. Numerical outcome had correlation with
the experimental results [14].
Adjusted delamination factor (Fda) was measured using a novel technique by Paulo
Davim, Rubio and Abrao et al (2007). Experimental design is also introduced for drilling
GFRP sheets with specific drilling parameters. Adjusted delamination factor (Fda) was
calculated by digital analysis. The digital analysis and experimental results have proved
that it is a good technique to estimate adjusted delamination factor (Fda) in drilling CFRP
[15].
Literature review
22
Tsao (2008) have found that the chisel edge of twist drill is the mainly influence for the
thrust force and the hole quality in drilling carbon fiber reinforced plastic (CFRP)
laminates. Pre-drilled pilot hole or reduce chisel edge can eliminate the threat for twist drill
in drilling induced-delamination. Drilling-induced thrust force was selected as quality
character factors to optimize the drilling parameters (drill type, feed rate and spindle speed)
to get the smaller the better machining characteristics by Taguchi method. The results
show that the feed rate and drill type are the most significant factor affecting the induced-
thrust in drilling CFRP laminates [26].
Palanikumar, Rubio, Abrao and Davim (2008) have proposed a mathematical model to
predict delamination in drilling glass fiber-reinforced plastic composites [16].
Oliver and Ekkard (2014) investigated low frequency vibration assisted drilling (LFVAD)
of CFRP/Ti6Al4 V [10/10 mm] in terms of tool wear and compared to conventional
drilling. Solid carbide drills with a diameter of 4.8 mm and different CVD and PVD
coatings have been tested. The flank wear as well as the adhesions at the cutting edges
have found to be significantly lower when using LFVAD. The tool life could be increased
by more than 300% compared to conventional drilling. This is based on considerably lower
process temperatures and an improvement of the process stability which could be proved
by cutting force measurements. Additionally the chip extraction was found to be more
efficient due to the generation of small chip segments which is a consequence of the
interrupted cut. Best results in terms of tool wear and borehole quality have been achieved
with an AlCrN coating [27].
2.3 Drilling parameters
All the outcome of drilling process i.e. thrust force, torque, delamination, wear of tool and
tool life etc. depends upon the cutting parameters (cutting speed and feed rate) and tool
materials and its geometry. Many researchers have studied the effects of tool parameters
like tool materials and geometry and cutting parameters like cutting speed and feed rate on
thrust force, torque and delamination. Their studies are important in predicting the
responses before the conduction of experiment so that the damages while drilling can be
reduced by selection of tool and cutting parameters.
Literature review
23
2.3.1 Effect of tool materials, types and its geometry on thrust force and torque
Tool geometries such as the point angle of the drill, drill diameter, helix angle, chisel edge
rake angle, web thickness have different impacts on the thrust force, torque and
delamination while drilling CFRP laminates.
Dharan and Won (2000) conducted drilling experiments in CFRP laminates of 9.9 mm
thickness and fiber volume fraction of 0.63 in the Matsuura MC510-VSS machining centre
using carbide-tipped twist drills and found that as the diameter of the drill is increased, the
thrust force and torque are also increased [28].
Mohan, Ramachandra and Kulkarni (2005) have conducted Drilling tests on glass fiber-
reinforced plastic composite (GFRP) laminates using CNC machining centre. Machining
parameters ( drill size, feed rate, and cutting speed) have been observed for damage-free
drilling of GFRP materials and developed a semi empirical relationship to characterise
drilling responses like thrust force and torque as functions of feed arte and speed by
conducting number of drilling experiments on glass fiber-reinforced polyester laminates
[25].
Velayudham and Krishnamurthy (2007) studied the influence of point geometry on thrust
and delamination. Drilling tests were carried out on glass fibre reinforced plastics using
carbide drills with different point geometries. Delamination is evaluated through ultrasonic
‘C’ scanning. The results shows that drill point has considerable influence on thrust and
delamination and tripod point geometry produce the least delamination damage [30].
Latha, Senthilkumar and Palanikumar (2011) analysed the influence of drill geometry on
thrust force in drilling GFRP composites. Three different drill bits namely, ‘Brad and Spur’
drill, ‘multifaceted’ drill, and ‘step’ drill in the experiment and response analysed is thrust
force and effect of its geometry on thrust force is studied. Three dimensional graphs are
used to analyse the results. Step drills are found better among the all drills under
consideration [31].
Satsangi (2012) developed surface roughness model for machining unidirectional glass
fiber reinforced plastics (UD-GFRP) composite using multiple regression methodology
and genetic algorithm approach. The experimentation was carried out with polycrystalline
diamond tool, covering a wide range of machining conditions. A second order
Literature review
24
mathematical model in terms of machining parameters was developed for predicting the
surface roughness using multiple regression methodology and optimized machining
parameters to minimize surface roughness [32].
Grilo, Paulo, Silva and Davim (2013) the influence of three distinct drill geometries and
cutting parameters (feed rate and spindle speed) in the delamination was assessed through
two delamination factors. A non-destructive method, based on processed images analyses
of the drilled surfaces, was used to measure the delaminated area and the maximum
diameter of damage zone. The best results were obtained with a Spur drill. With this, the
higher rate of production, without the occurrence of delamination, was obtained with a
feed rate of 2025 mm/min and a spindle speed of 6750 rpm [33].
Karpat and Bahtiyar (2015) have used a systematic approach to compare the influence of
drill geometry on process outputs such as drilling forces, torques and tool wear. Custom-
made double point angle polycrystalline diamond (PCD) drills from the same manufacturer
were used in the experiments. The advantage of this approach is that it eliminates the drill
material and edge preparation effects on the experimental measurements, thus helps reveal
the influence of drill geometry on the process outputs. The pros and cons of different drill
designs are discussed and an appropriate design is identified for the drilling of thick CFRP
laminate considered in this study [34].
2.3.2 Effect of tool materials and its geometry on delamination
Delamination can be the responsible factor for limiting the use of composite materials in
structural applications because it decreases the stiffness and strength of a composite plate
and load carrying capacity as well, particularly when compressive, shear and fatigue type
of loads are applied and when exposed to moisture and tuff environments for a for
prolonged time.
Piquet et al. (2000) analysed the effects of drilling tool geometry on the drilling quality of
thin carbon/epoxy plates and found that for a conventional double fluted twist drill to give
good results on these plates, it is necessary to pre-drill a hole in order to neutralize the
chisel edge effect and to lubricate the machining process. Machining conditions can further
be improved by applying a variable feed rate in relation to its geometry [11].
Literature review
25
Tsao (2008) performed drilling experiments in CFRP laminates using twist drills of
different geometries and found that reducing point angle of the twist drill results in
substantial increase in relief angle and decrease in thrust force. They also concluded that
carefully selected drill geometry and small feed rate produces low thrust force in drilling
which can reduce the threat of induced delamination while drilling using twist drill [26].
Durao et al. (2010) investigated the performance of five tungsten carbide drills of 6 mm
diameter and different geometries such as twist drill with a point angle of 120°, a twist drill
with a point angle of 85°, a brad drill, a dagger drill and a special step in drilling CFRP
laminate and found that twist drill with 120° point angle has always the highest force but
the delamination is minimum [35].
Palanikumar et al. (2011) investigated the influence of twist drills of different point angles
such as 85°, 115°and 130°in drilling glass/epoxy composite material. The results between
the different drill point angles indicate that the 85° point angle gives better results than
115° and 130° and therefore, they have concluded that the drill with 85° point angle
produces less delamination than the drills with 115° and 130° point angles [36].
2.3.3 Effect of cutting parameters (speed and feed rate) on thrust force and torque
From the literature survey it is found that with increase in speed and feed rate thrust force
increases especially with increase in feed rate because larger the feed rate the cross
sectional area of unreformed chip will be higher, so more resistance to chip formation
which in turn result in greater axial thrust force.
Lin and Chen (1996) performed high speed drilling of CFRP laminates using tungsten
carbide twist drill and multifacet drill and found that as the spindle speed was increased
from 9550 rpm to 38650 rpm (from 250 to 850 m/min) (Ø7 mm) both thrust force and
torque increased. Although tools were worn out quickly and the thrust force increases
drastically as cutting speed increases, an acceptable hole entry and exit quality was
maintained. This was because relatively small feed rates were used in these tests [37].
Dharan and Won (2000) used tipped carbide twist drill to perform drilling experiments in
Carbon Fibre Reinforced Plastics. They found that with increase in feed rate thrust force
and torque also increased (from 100mm/min to 1000mm/min). They used laminates of
9.9mm thickness [28].
Literature review
26
Mohan, Ramachandra and Kulkarni (2005) conducted series of experiments and on Glass
Fibre Reinforced Plastics using CNC milling centre and found that torque and thrust force
are the functions of drill size and feed arte and they have developed an empirical relation
which models the cutting speed and feed rate with response of thrust force and torque.
They established that empirical relation co relates better for small size drill at lower feed
ranges [25].
Fernandes and Cook (2006) performed an experimental study on drilling carbon
composites using a special type of drill tool (one shot drill) and found that thrust force is
increased with increase in feed and tool wear [38].
Zitoune and Collombet (2007) have suggested a Numerical finite element analysis model
which was function of tool point geometry and the shear force effects in the composite and
compared it with analytical models to calculate the thrust force at exit of the drilled hole in
FRP laminates. They have validated the numerical model on two types of products of
carbon-epoxy materials by conducting punching experiments at low speeds. Numerical
outcome co related with the experimented results [14].
Tsao and Hocheng (2007) carried out drilling experiments on CFRP laminates using core
drill 10mm in diameter fitted with diamond at front end and twist drill. They took drill
thickness, grit size, feed rate and spindle speed as parameters and found that the grit size
and feed rate are the main effective parameters. The thrust force of core drill and twist drill
increases with increase in feed rate. The spindle speed was relatively insignificant
parameters [39].
Tsao (2008) conducted experiments on drilling CFRP composites using step core drills and
found that the thrust force decreases significantly with an increase in spindle speed from
800 to 1200 rpm. The thrust force of various step-core drills increases with decrease in
diameter ratio and increase in feed rate [26].
Jayabal and Natarajan (2010) have developed mathematical model to correlate the effects
of cutting parameters and their responses and also found out the optimum values of cutting
parameters to minimise the thrust force and torque by conducting experiments [40].
Rahamathullah and Shunmugam (2011) experimented micro drilling of glass-fibre-
reinforced plastic (GFRP) using a carbide drill of 0.32 mm diameter. Micro drilling
Literature review
27
experiments on GFRP have been carried out using a full-factorial design with five levels
for speed and feed rate and three repetitions for each run. For the purposes of comparison,
micro drilling of plain sheet made out of matrix material has also been carried out. The
experiments on blind-hole drilling reveal that there is a reduction in maximum thrust force
and torque in peck drilling. Encouraged by this, with peck cycle, through-hole drilling of
2.25 mm thick GFRP specimens has been carried out successfully. Regression models
developed for the thrust force show good correlation between the measured and predicted
values, while the torque values lie scattered with respect to the predicted trends. The results
indicate that micro-holes of large aspect ratio could be produced by selecting proper
process parameters and drilling strategy [41].
Vankanti and Ganta (2013) carried out experiments as per the Taguchi experimental design
and an L9 orthogonal array was used to study the influence of various combinations of
process parameters on the quality of hole. They found that feed rate is the most significant
factor which affect the thrust force followed by speed, chisel edge width and point angle;
cutting speed is the most significant factor affecting the torque, speed and the circularity of
the hole followed by feed, chisel edge width and point angle [42].
Su, Wang, Yuan, and Cheng (2015) have established theoretical model of drilling of the
drill tapered reamer to study push out delamination and analysed thrust force and
delamination. Thrust force increases with increase in feed rate. To improve the hole
quality, the length of the cutting edges of the tapered drill-reamer should be about 9.6 mm,
and the improved drilling method of placing two rigid plates on both sides of CFRPs
workpiece is preferred [43].
2.3.4 Effect of cutting parameters (speed and feed rate) on delamination
Tagliaferri et al (1990) developed a novel method to measure the width of the damage zone
in drilling of GFRP. The delamination zone is correlated to the ratio between the drilling
speed and feed rate. Higher the ratio better is surface roughness. However, the damage
zone decreases for a definite ratio beyond which damage stays constant [44].
Chen (1997) identified that the effect of cutting speed on cutting forces differs with various
tool materials, but it seems that there is no effect for the same drill material. In metal
cutting, the effect of cutting speed on the work-hardening of work material is eliminated by
Literature review
28
the softening of the material due to the increasing cutting temperature. This phenomenon
may also be found in the drilling of CFRP composite materials. The built-up edge is not
found in the drilling of CFRP composite materials. However, the effect of cutting speed on
the cutting force is insignificant for the same drill material. It can be found that the lower
the feed rate, lower is thrust force and torque generated in drilling. In order to improve the
hole quality the feed rate needs to be decreased during the drilling process. However, if the
feed rate is too low, the cutting time at the same place is too long, and thus the
delamination easily occurs owing to the deviation affected by vibration in the high spindle
speed [45].
Edoardo Capello (2004) analysed the difference between delamination mechanisms when
drilling with and without a support placed under the workpiece. They designed a new
device that counters the hypothesized delamination mechanism and built a prototype of this
device and its effectiveness verified. They derived that the proposed device can drastically
reduce delamination [23].
Hocheng and Tsao (2005) derived the path towards delamination-free drilling of composite
materials. The used drill bits like step drilling, pilot hole, back-up plate and various non-
traditional methods have been reviewed. The different drills show different level of the
drilling thrust force varying with the feed rate. The special drill bits can be operated at
larger feed rate without delamination damage compared to the twist drill [46].
Davim, Rubio and Abrao (2007) have given experimental design for drilling FRP
laminates under particular cutting conditions. They have digitally analysed the damage for
the assessment of delamination factor. The results indicated that the digital analysis can be
suitably used to estimate the damages produced after drilling carbon fibre reinforced
plastics (CFRP) [15].
Gaitonde, Karnik and Davim (2008) have used the utility concept for multi-performance
characteristics optimization using Taguchi design. They have carried out experiments as
per L9 orthogonal array different conditions of feed rate and cutting speed in each
experiments. The Analysis of means (ANOM) and Analysis of variance (ANOVA) were
performed to determine the optimal levels of the parameters and to identify the level of
importance of the machining parameters on delamination factor respectively [47].
Literature review
29
Fotouhi, Pashmforoush, Ahmadi and Oskouei (2011) gave a method using acoustic
emission feature for delamination free drilling in glass-epoxy composite material. Three
technics were used to monitor acoustic emission i.e. sentry function, acoustic emission
energy distribution, and acoustic emission count distribution. They developed a technic to
find thrust force at the onset of delamination using three point bending tests and simulated
thrust force in drilling without using back plate. Sentry function is found as result to
combine AE information and mechanical behaviour of composite materials. Specimen
used were having two different lay ups woven [0, 90]s and unidirectional [0]s, leading to
different levels of damage evolution. Results show that AE parameters and sentry function
method are useful tools for the examination of initiation and the growth of delamination
during drilling process and can help to avoid delamination damage while drilling [48].
Kilickap (2011) presented a comprehensive mathematical model for correlating the
interactive and higher order influences of drilling parameters on the delamination factor in
drilling glass fiber reinforced plastic (GFRP) composites using response surface
methodology. They investigated the influence of drilling parameters, such as cutting speed,
feed, and point angle on delamination produced when drilling GFRP composite [49].
Khaled Giasin and Sabino Ayvar (2016) performed an experimental study to analyse the
effects of drilling parameters (spindle speed and feed rate) on quality of hole in two grades
of GLARE (2B & 3). They evaluated the hole size, circularity error, entry and exit burrs,
chip formations and damage described at the macro level (delamination area) using
computerised tomography CT scan, and at the micro level (fibre matrix de-bonding,
chipping, adhesions, cracks) using scanning electron microscopy (SEM). They statistically
analysed using analysis of variance (ANOVA) to determine the contribution of cutting
parameters on investigated quality of hole and its parameters [50].
2.3.5 Effect of combination of tool and process parameters
Konig et al (1985) studied tool geometries and cutting conditions in machining of FRPs
and found that tool geometry with protruding peripheral cutting edges can reduce the thrust
force. Water jet cutting technic is suitable for thin laminates, but requires careful
adjustment of the cutting parameters in order to avoid delamination and chipping at the jet
exit side [51].
Literature review
30
Caprino and Tagliaferri (1995) studied intermittent drilling by checking the development
of cracks at regular intervals and pre-set depths by metallographic technique and
examining by optical microscopy. Step-wise delamination, inter-laminar cracks and high
density micro failure zones were monitored when the feed rate is high and established a
relationship between the microscopic examination and macroscopic damage evaluated by
visual examination [52].
Langella et al (2005) presented a mechanistic model for predicting torque and thrust during
drilling of GFRP materials. The influence of feed rates and point angles on thrust and
torque was studied, distinguishing between the respective contributions of the cutting lip
and the chisel edge. The total force generated by cutting lip and chisel edge has shown that
the thrust force and the damage of the material increase proportionately. The effect of
chisel edge on thrust force increases to the feed rate and may account for over 80% of the
total force needed to drill a hole [53].
Tsao (2006) experimentally obtained the thrust force and surface roughness of core drill
with respect to various tool and process parameters (grit size of diamond, thickness, feed
rate and spindle speed) in drilling of CFRP laminate. The experimental results indicate that
thickness and feed rate are recognized to make the most significant contribution to the
overall performance. For thrust force, the thickness and feed rate are the most significant
factors, whereas for surface roughness, it is the feed rate and the spindle speed [54].
Tsao (2008) studied the influence of twist drill geometry in drilling of CFRP composite
material. The point angle, helix angle and relief angle were varied and three different drill
types were used in the experimentation. Taguchi analysis was carried out with three
factors, viz. drill type, feed rate and spindle speed. The analysis of thrust force and signal-
to-noise ratio indicates that the feed rate and drill type are the main parameters among the
three control factors that influence the thrust force. The effect of spindle speed was
relatively insignificant [26].
Krishnaraj (2008) studied the effects of drill points on GFRP while drilling at high spindle
speed. Drilling experiments were conducted with twist drill, Zhirov-point drill and
multifaceted drill with wide range of speed and feed to analyse thrust force, delamination
and surface roughness. At high speed, thrust force is less and further the Zhirov-point drill
Literature review
31
improves the quality of hole. Multifaceted was found suitable to minimize delamination
[55].
Durao et al (2010) studied the drilling process of composite laminates with respect to
cutting parameters, tool material and geometry. The use of HSS, tungsten carbide (WC)
and PCD drills with the geometry of twist, Brad and step were studied [56].
2.4 Modelling of Drilling Parameters
Researchers have studied many modelling technics to minimize the delamination and to get
the values of parameters for delamination free drilling before actual experiment. Modelling
can be achieved by feeding the set (training data) of parameters to the developed models
that accurately predicts the responses for any other set of machining parameters. Some of
the modelling technics are discussed here from the references.
2.4.1 Response Surface Methodology
Response surface methodology (RSM) explores the relationships between
several explanatory variables and one or more response variables. The main idea of RSM
is to use a sequence of designed experiments to obtain an optimal response. Box and
Wilson suggest using a second-degree polynomial model to do this. They acknowledge
that this model is only an approximation, but they use it because such a model is easy to
estimate and apply, even when little is known about the process. Statistical approaches
such as RSM can be employed to maximize the production of a special substance by
optimization of operational factors. In contrast to conventional methods, the interaction
among process variables can be determined by statistical techniques.
Wang et al (1995) reported orthogonal cutting mechanisms of multidirectional composite
laminates using PCD tools. Experiments were executed following a full factorial
experimental design based on RSM technique. Cutting force measurements and chip
formation were recorded using a CCD camera. Post process measurements included
surface profilometry and SEM of the machined surfaces [57].
George (2004) implemented RSM technique in electric discharge machining of carbon-
carbon composite plate and established the relationship of Empirical models correlating
process variables and their interactions with the response functions. RSM model can be
Literature review
32
used for selecting the values of process variables to get the desired values of the response
parameters [58].
Gaitonde et al (2008) investigated the effect of parametric influence on delamination in
high-speed drilling on CFRP. Experiments were performed based on full factorial method
of Taguchi design of experiment by selecting three levels of each and response values of
delamination factor were empirically related to process parameters by developing a second
order nonlinear regression model based on RSM [59].
Kishore, Tiwari, Rakesh, Singh and Bhatnagar (2011) investigated UD-GFRP laminates in
drilling the response surface methodology and established the optimum levels of geometry
(cutting speed and feed rate) for minimizing the damage in drilling GFRP laminates. They
also found that the delamination and other damages have significant influence of tool
geometry [60].
Habib, Patwari, Jabed and Bhuiyan (2016) have developed a hybrid model of Harmony
Search (HS) with Response Surface Methodology (RSM for optimizing the surface
roughness of three different GFRP composite materials during drilling operation [61].
Kilickap (2016) used response surface methodology and developed a mathematical model
to correlate the interactive and higher order influences of drilling parameters on the
delamination factor [62].
2.4.2 Fuzzy Logic
The classes of certain objects in the real world do not have precisely defined criteria of
membership. Fuzzy set was introduced by Zadeh (1965) to deal such problems and is
defined as a class of objects with a continuum of grades of membership. The fuzzy set
assigns a grade of membership that ranges between 0 and 1 to each object replacing
linguistic variables. Rule base is framed based on if-then rules. This fuzzification of data is
then defuzzified by aggregation of these rules and converting the fuzzy quantity to a
precise quantity [63].
Karthikeyan et al (2002) used fuzzy logic and genetic algorithms to optimize the drilling
characteristics for aluminium composites. They studied drill wear, specific energy and
surface roughness and the process parameters considered were volume fraction, cutting
Literature review
33
speed and feed rate. Fuzzy logic is used to train and simulate experimental data and
optimization of cutting parameters were performed using genetic algorithms and validated
experimentally [64].
Yaldiz et al (2006) predicted cutting forces using fuzzy model which were obtained by
designed dynamometer in turning. As inference system A Mamdani max-min method was
used and for defuzzification centroid method is used. The difference between predicted and
experimental results was obtained as around 99.6% [65].
Latha and Senthilkumal (2009) predicted thrust force in drilling of composite materials
using fuzzy logic. Fuzzy rule-based model was developed to predict the thrust force in
drilling of GFRP composites. Fuzzy model and the response surface model were
compared. Accurate results were achieved between the predictive model values and
experimental values [66].
Vimal (2009) developed fuzzy rule based model to predict thrust force and torque in
drilling of GFRP. Their fuzzy based model can be effectively used for predicting the
response variable by means of which delamination can be controlled. verification tests
were conducted for the confirmation of the fuzzy logic rule based modelling which
indicates that difference between the experimental values and the predicted values were
very small so fuzzy rule based modelling technique is effective for the prediction of thrust
force and torque in drilling of GFRP composites [67].
2.4.3 ANFIS (Adaptive Neuro Fuzzy Inference System)
Complex uncertain real world problems can be effectively modelled by fuzzy inference
system (FIS) incorporating qualitative aspects of human knowledge and reasoning process,
without employing a precise quantitative analysis. Fuzzy logic rules and membership
functions are used in the form of knowledge, which is collected from the experimental
results, to approximate the expert perception and judgment in modelling the process input-
output relationship using linguistic variables rather than a complicated dynamic model.
Inference operations are performed on the logic rules using operators within the rules. The
fuzzification interface transforms the crisp inputs into degrees of match with linguistic
values and the fuzzy inference results into crisp output is converted by defuzzification
interface.
Literature review
34
ANFIS was first introduced by Jang (1993) for a novel architecture which can be used as a
basis for constructing a set of fuzzy if then rules with appropriate membership functions to
generate the stipulated input-output pairs. Nonlinear functions can be modelled by ANFIS
architectures. Comparisons with Artificial Neural Networks and work on fuzzy modelling
were mentioned which shows that the error index is the least in ANFIS. It was suggested
that the effective partition of the input space can decrease the number of rules and thus
increases the speed in learning and application phases.
Tsai and Wang (2001) compared the MRR of the work pieces for the different materials
considering the change of polarity among six different neural networks together with a
neuro-fuzzy network. Same experimental data is used to train all these six different neural
network. A comparison of error showed that ANFIS with bell-shape membership function
is the best model [68].
Chinnam and Baurah (2004) utilized a neuro-fluffy methodology for assessing mean
residual life in condition-based maintenance frameworks. They utilized ANFIS monitor
high-speed-steel drill bits used for drilling holes in stainless steel metal plates. A fuzzy
inference model is built up failure definition and is evaluated and is assessed on a cutting
device monitoring problem. The plots of membership functions before and after training
for thrust force and torque are accommodated better understanding of FIS model [69].
Samhouri and Surgenor (2005) experimented an online monitoring and prediction of
surface roughness in grinding using ANFIS. The model used a piezoelectric accelerometer
to produce a signal related to grinding features and surface finish. The power spectral
density (PSD) of this signal is utilised as an input to ANFIS model, which responded a
value for the online monitoring and prediction accuracy of surface roughness. Validation
by experiments showed a good agreement of the measured surface roughness with the
predicted values. The adoption of bell-shaped functions achieved a satisfactory online
prediction accuracy of 91% [70].
Jagdev and Simranpreet (2009) utilised ANFIS model for simulation of ultrasonic drilling
of porcelain ceramic with hollow stainless steel tools. Two input signals were depth of
penetration and time for penetration and output signal was MRR. The validation
experiment showed that the predicted values are well in agreement with the experimental
values at 0.1% level of significance [71].
Literature review
35
2.5 Optimization
The quality of a product depends on the parameters involved in the manufacturing process.
For Optimization of a manufacturing process one has to consider all the factors involved
which can influence the quality of product and productivity. There are so many techniques
available for optimisation and researchers have used different techniques for different
solutions.
2.5.1 Grey relational analysis
The theory of grey systems is a new technique for performance prediction, relational
analysis and decision making in many areas. In a system that is complex and multi-
variable, the relationship between various factors is unclear. Such systems are often
referred as “grey” implying poor, incomplete and uncertain information. Their analysis by
classical statistical procedures may not be acceptable or reliable without large datasets that
satisfy certain mathematical criteria. But, the grey theory uses relatively small datasets and
does not demand strict compliance to certain statistical laws.
George et al (2004) determined the optimal setting of the process parameters on EDM
machine while machining carbon-carbon composites using a Taguchi approach based on
RSM and ANOVA. It was found that electrode wear rate reduces substantially within the
region of the experimentation, if the parameters are set at their lowest values, while the
parameters set at their highest values increase the MRR drastically [72].
Xie et al (2007) used grey relational analysis for optimizing the square hole flanging
process parameters in sheet metal forming with considerations of the multiple responses.
Grey relational grade is obtained and ANOVA is applied which showed good agreement
with the experimental results [73].
Chang and Lu (2007) applied a grey relational analysis to a set of 2 stage experiments
designed to determine the cutting parameters for optimizing the side milling process with
multiple performance characteristics. It is found that his approach is simple and efficient in
determining an optimal combination of the cutting parameters. The confirmation tests also
show that this approach can improve the cutting process [74].
Literature review
36
Noorul Haq et al (2008) presented a multi response optimization of machining parameters
in drilling MMC using grey relational analysis. Drilling tests were carried out using TiN
coated HSS twist drills of 10 mm diameter under dry conditions. The process parameters
chosen were speed, feed and point angle and the multi responses were surface roughness,
cutting force and torque. Based on the grey relational grade, optimum levels of parameters
had been identified and significant contribution of parameters was determined using
ANOVA. Confirmation test was conducted to validate the test results [75].
Pal et al (2009) optimized the quality characteristics parameters in a pulsed metal inert gas
welding process using grey-based Taguchi method. Many quality characteristic parameters
are combined into one integrated parameter by using grey relational grade. ANOVA has
been performed to find the impact of individual process parameter on the quality
parameters. Validation of the results has confirmed the effectiveness of grey relational
analysis for optimization of the chosen manufacturing process [76].
Kurt et al (2009) utilized Taguchi method and grey relational analysis to optimize surface
finish and hole diameter accuracy in the dry drilling of aluminium alloy. The parameters of
hole quality are analysed under varying cutting speed, feed, depth of drilling and different
types of drilling tools under dry drilling conditions. Confirmation tests with the optimal
levels of machining parameters are carried out to illustrate the effectiveness of the grey
relational approach [77].
Jailani et al (2009) optimized the sintering parameters of aluminium alloy using grey
relational analysis. Al-Si alloy powder was homogeneously mixed with various weight
percentages of fly ash and compacted at varying pressures. Optimal levels of parameters
were identified using grey relational analysis and significant parameter was identified
using ANOVA [78].
Experimental results indicate that multi-response characteristics such as density and
hardness can be improved through grey relational analysis.
Ku et al (2010) developed a new type of thermal friction drill made of sintered carbide and
conducted the experiments on stainless steel plate using Taguchi method. The optimization
of the drilling parameters with respect to surface roughness and bushing length using
ANOVA was carried out and finally the confirmation experiments showed that friction
angle and speed are the two significant parameters that affected surface roughness [79].
Literature review
37
2.5.2 Grey fuzzy approach
The grey relational analysis is an improvement of grey relational analysis with the
implementation of fuzzy logic theory in the multivariate system to obtain better system
performance.
Lin et al (2002) optimized the EDM process based on the orthogonal array with fuzzy logic
and grey relational analysis method. It was concluded that grey relational analysis is more
straightforward than the fuzzy-based Taguchi method for optimizing the EDM process
with multiple process responses [80].
Tosun (2006) used grey relational analysis for optimizing the drilling process parameters
for the work piece surface roughness and the burr height. An orthogonal array was used for
the experimental design. Speed, feed, drill and point angle were considered as input
drilling parameters that need to be optimized for obtaining multi-performance
characteristics [81].
Chiang et al (2008) investigated optimal machining parameters for the die casting process
of magnesium alloy using grey based fuzzy algorithm. Experimental results have shown
that the required performance characteristics in the die casting process have great
improvements by using this proposed algorithm [82].
A method of grey-fuzzy approach was proposed by Liu et al (2009) to achieve
optimization of multi-response characteristics during the manufacturing process of the light
guide plate printing process. The experimental results using the optimal setting improved
the manufacturing process in this study [83].
Montesano, Bougherara and Zouheir (2017) have experimented the effects abrasive water
jet and conventional drilling on CFRP and found that abrasive water jet (AWJ) is better in
drilling the CFRP plates for fatigue properties [84].
Khaled, Sabino, French and Phadnis (2017) have investigated the cutting forces induced at
different cutting speed and feed rate and modelled it using 3D finite element modelling on
Glass fibre aluminium reinforced epoxy (GLARE) [85].
Literature review
38
Jasvir Singh and Ashwani Kumar (2018) have studied the effects of cutting speed, feed
rate, thickness of plates using Taguchi’s L27 array and ANOVA analysis and found the
feed rate as most responsible factor [86].
2.6 Motivation for the Research
Glass Fibre Reinforced Plastics are widely used in industries due to its properties like
strength to weight ratio. It is essential to study because it is emerging in all domains of
engineering. Glass Fibre Reinforced Plastics composite material is widely used composite
materials in all such domains. Composites are made near to shape by moulding and other
engineering technics which are based on the requirements but assembling of parts is
mandatory for desired shape and structure of dimensional accuracy. Drilling is unavoidable
operation for joining the parts. From the above study it is clear that during drilling of
GFRP, there are variety of damages encountered, viz. delamination, and fiber pull-out, de-
bonding and cracking.
These damages are produced mainly due to the following factors:
(1) Structural characteristics of the Glass fiber, the type of the matrix used, and the
stacking sequence used in the formation of CFRP composites.
(2) Material of the tool used and the tool geometry.
(3) Cutting parameters i.e. speed, Feed rate that control the quality of holes produced.
Researchers have investigated the factors affecting the quality of holes drilled in the Glass
fiber composite laminates. An in-depth review on the literatures dealing with the drilling
factors such as cutting parameters, tool geometries, tool types and materials influencing the
quality of drilled holes in composite laminates was carried out. Further, literature dealing
with the defects such as delamination, surface roughness, roundness error and tiny cracks
in the hole of wall which occurs during the drilling of fiber reinforced laminates was
studied. Few researchers have emphasized the need for online monitoring the drilling of
polymeric composite laminates owing to the growth in the field of intelligent machining
and increase in productivity. Among all the online monitoring techniques available,
methods such as monitoring thrust force, torque and acoustic emission are widely
employed by the researchers due to their high sensitivity to changes in the process
parameters and to the intensity of the process damages. Literature on Fuzzy models and
Literature review
39
ANN models employed for predicting the outcome of the drilling process has also been
extensively reviewed.
However, in Glass Fibre Reinforced Plastic composites the application and comparison of
the tools and its geometry including Twist drills, Step drills and multifaceted drills with
their point angles are not reported much. There is no systematic and comprehensive study
of optimization of drilling parameters for drilling of GFRP composites. The identification
of these factors motivated for this research work to study the damages caused in the form
of delamination, eccentricity and surface roughness using Twist drills, Step drills and
multifaceted drills.
2.7 Problem Formulation
From the literature review it is self-explanatory that Drilling is an important operation in
assembling the GFRP parts and the quality of holes depends upon cutting parameters like
cutting speed, feed rates, tool types and its geometry, material of tools etc. Thrust force of
cutting is the result of combination of all above parameters which is a major responsible
factor for the defect called delamination in drilling. With proper selection of cutting
parameters and tool geometry, how and up to what extent delamination can be controlled
or eliminated is a major focus of this study.
From the above motivation, the problem defined is to study and analyse the major damages
in drilling of Glass Fibre Reinforced Plastic composite material which is delamination.
Using different tool geometry the analysis is done and the best tool and geometry is to be
selected for industries. At the same time cutting parameters for delamination free drilling is
also need to be suggested for the particular GFRP composites. The experimental work is to
be designed and statistical model need to be suggested. As the optimization of the cutting
parameters to minimise delamination is not emphasized fully in the GFRP drilling with
different tool geometry, it needs to be addressed. 0So it is decided to implement the
combination of fuzzy model and artificial neural network which is known as ANFIS
(Adaptive Neuro Fuzzy Inference System) and to create a model in MATLAB so that as
and when required one can select optimum cutting parameters for delamination free
drilling.
Literature review
40
2.8 Scope and Objectives
In this research, the problem is defined as an experimental investigation, Monitoring and
control, modelling and optimization of drilling parameters in drilling of GFRP laminates.
There is vast scope in the aerospace industry because it is investigated that more than 60%
rejection of materials at assembling stage in aerospace industry is due to delamination so
monitoring and control of delamination and a model to select the cutting parameter for
delamination free drilling can be very much useful and can save money and manpower
both.
2.8.1 Objectives
(1) To Monitor the occurrence of delamination in drilling and to observe the thrust
force and torque with respect to cutting speed and feed at the onset of delamination
in a specially manufactured GFRP having specific mechanical properties.
(2) To investigate the effect of different tool geometry and cutting parameters on the
delamination factor.
(3) To develop mathematical model which can be a readymade tool to select the
cutting parameters and point angles of tools for delamination free drilling within a
given range.
2.8.2 Methodology
The methodology of the proposed research to be carried out is shown in Figure 2.1
Literature review
41
FIGURE 2.1 Methodology
Monitoring and control of delamination in drilling of GFRP
Drilling of GFRP Laminates on VMC
Deciding Machining variables
Offline measurements 1. Delamination Factor at
entry 2. Delamination Factor at exit
Speed
(rpm) 1500 2000 2500
Point angle (degree)
118 130 140
Feed (mm/min)
100 200 300
Monitoring the drilling process
Online measurements 1. Thrust force (N)
2. Torque (N.m) (With Kistler 9257 Piezo Electric Dynamometer)
Fabrication of GFRP laminates
Data collection and Analysis
Results and Discussion Effect of cutting parameters on
1. D.F. at entry 2. D.F. at exit
3. Thrust Force and Torque Modelling and validation using ANOVA and ANFIS in
MATLAB
Conclusion and future scope
Experimental Work
42
CHAPTER – 3
Experimental Work
Glass fibre reinforced plastics are having desirable properties against conventional
materials such as high specific strength and modulus, variable directional strength
properties and better fatigue strength. Due to its above properties they are widely used in
aerospace and aeronautical industries. Drilling is an inevitable process to be carried out in
composites for the assembling of parts. The drilling of GFRP composite materials is
completely a different approach when compared to the conventional materials. Certain
defects like delamination at entrance, Delamination at exit, fibre pull out, spalling and fibre
breakage are encountered while drilling of GFRP. In depth study of drilling is required to
investigate the factors responsible for above and defects and to reduce it up to the
acceptable limit. This chapter includes the work material (GFRP) along with its properties,
specifications and fabrication of GFRP composite laminates. The different tools used in
this study along with its point angle and geometry are presented. The various cutting
parameters for drilling and the responses are explained. The schematic and practical
experimental setup and the experimental procedure are described. Finally, the experimental
results obtained for different tool at respective cutting parameters are tabulated.
3.1 Specimen Fabrication and Material Properties
Glass fibre reinforced plastic (GFRP) composite material is used for the experimentation
purpose in the present study. GFRP is having enormous application in industries now a
days due to its characteristics of higher strength to weight ratio, light weight, corrosion
resistance, low thermal conductivity, resistance to environment, low thermal expansion etc.
Apart from glass fibre and resin additionally additives and fillers are added in the
composites to achieve the above properties in the material. Design considerations for the
fabrication of specimen includes type of fibre reinforcement, orientation of fibre and its
percentage in weight fraction and type of resin. In the present research work E-glass fibres
are used as a reinforcement of 9 micron diameter and epoxy is used as a resin. Weight of
Experimental Work
43
fibre is 70 % by proportion and placed unidirectional through the material with continuous
fibres. Epoxy resins have superior mechanical properties, resistance to corrosive liquids
and environments, superior electrical and thermal properties, and mainly good adhesion to
the substrate. Proper selection of hardeners is done to control Cure rates. The epoxy resins
used in the present work is ARALDITE LY556 mixed with the HARDENER XY 54 with
the proper mixing ratio. The resin and the hardener are mixed uniformly until they form a
homogeneous mixture. In the present research work composite laminates of GFRP are
manufactured by vacuum infusion process at manufacturing facility of ATIRA
(Ahmedabad Textile and Industrial Research Association).
3.1.1 Fabrication
Vacuum infusion is useful to produces high and consistent level of laminate quality and it
produces very less amount of hazardous air pollutants. Vacuum infusion process uses
vacuum pressure to drive resin into a laminate. The reinforcements, in this case glass
fibres, are arranged in a mould and covered with a flexible bag. The boundary is sealed and
then vacuum is applied which draws resin from a container into the bag along a specified
direction.
FIGURE 3.1 Vacuum Infusion Process
Experimental Work
44
Advantages:
(1) Better fiber-to-resin ratio can be accurately achieved with low void contents.
(2) Less amount of hazardous air pollutants Good health and safety.
(3) Very consistent resin usage.
GFRP Laminates were made in the size of 300mm *300 mm.
3.1.2 Testing
Laminates were also tested for the required properties at ATIRA testing laboratory. Tests
were performed for tensile strength and tensile modulus, compressive strength and
modulus and hardness of material. Dual column floor mounted Electromechanical Testing
System made by INSTRON Company is used for testing the properties. Five samples of
standard specimen are taken from the fabricated laminates of GFRP for each test and
average value is taken for the experiment purpose. Standard samples and testing are shown
in the Figure 3.2 (ISO 527-4 is used for testing while ASTM standard is D638 for tensile
test) and Figure 3.3. Each specimen is fixed in the electro mechanical jaws after special
preparation and the load is increased gradually so that first sign of crack develop in the
specimen. Automatically that load is recorded in the computer and graph for load verses
extension in length is generated which is shown in the Figure 3.4(a). Figure 3.4(b) shows
the report of compressive test.
(a) Actual specimen (b) Specimen geometry
FIGURE 3.2 Standard specimen for tensile test
Experimental Work
45
Testing Machine specification:
(1) 400 KN capacity
(2) 2050 mm (80.7 in) vertical test space
(3) Load measurement accuracy: +/- 0.5% of reading down to 1/1000 of load cell
capacity option (2580 Series load cells)
(4) Up to 2.5 kHz data acquisition rate option simultaneous on load, extension, and
strain channels
(5) Speed range of 0.00005 to 1016 mm/min (0.000002 in/min to 40in/min)
Hardness of laminates were tested by BARCOL hardness tester and Figure 3.4(c) shows
the hardness test report. All the test reports were done and approved by ATIRA material
testing laboratory.
Mechanical properties of the GFRP composite laminates are shown in Table 3.1
TABLE 3.1 Mechanical properties of GFRP laminates
Name of Properties value
Tensile strength 732 MPa
Compressive strength 693 MPa
Tensile Modulus 46 GPa
Compressive modulus 33 GPa
Density 2.4 g/cm3
Thickness 4 mm
Woven weaving base E Glass % by volume 70 %
Diameter of glass fibre 9 µm
Epoxy Resin % by volume 30 %
Barcol Hardness 70 BHN
Experimental Work
46
FIGURE 3.3 Specimen Testing for Tensile strength
Experimental Work
47
FIGURE 3.4 (a) Tensile test report
Experimental Work
48
FIGURE 3.4 (b) Compressive test report
Experimental Work
49
FIGURE 3.4 (c) BARCOL Hardness test report
Experimental Work
50
3.2 Selection of Tool and Tool Geometry
Drill tools are used to make holes in the GFRP composite laminates. Selection of a
particular tool and its geometry inn context with cutting parameters significantly affect the
responses of the process i.e. thrust force and delamination.as discussed in literature review
tool geometry and in particular point angle have great influence on thrust force and torque
and so on delamination at exit and entrance.
Palanikumar et al. (2008) investigated the influence of twist drills of different point angles
such as 85°, 115°and 130°in drilling glass/epoxy composite material [16]. The results
between the different drill point angles indicate that the 85° point angle gives better results
than 115° and 130° and therefore, they have concluded that the drill with 85° point angle
produces less delamination than the drills with 115° and 130° point angles.
Tsao (2008) performed drilling experiments in CFRP laminates using twist drills of
different geometries and found that reducing point angle of the twist drill results in
substantial increase in relief angle and decrease in thrust force [26]. They also concluded
that carefully selected drill geometry and small feed rate produces low thrust force in
drilling which can reduce the threat of induced delamination while drilling using twist drill.
Durao et al. (2010) selected five drills of tungsten carbide with 6 mm diameter and
different geometries such as twist drill of point angle of 120°, twist drill of point angle of
85°, dagger drill, special step drill and brad drill to drill CFRP laminate and found that
point angle 120° and having twist drill geometry has the highest force but the delamination
is minimum [35].
However, in Glass Fibre Reinforced Plastic composites the application and comparison of
the tools and its geometry including Twist drills, Step drills and multifaceted drills with
their different point angles are not reported much.so it is decided to select the above tool
geometries with point angle of 1400 of twist drill, 1300 of step drill and 1180 of
multifaceted drill which are not readily available commercially so all these tool are
manufactured at Kreative Tooling’s, Vatva GIDC, Ahmedabad from High Speed
Steel(HSS) and titanium nitride coatings. Nine types of each and total 27 tools were
manufactured. Figure 3.5 (a), Figure 3.5 (b) and Figure 3.5 (c) demonstrates different tools.
Other details of tool geometry for all tools are tabulated in Table 3.2.
Experimental Work
51
Table 3.2 Tool and its geometry details
Twist drill Step drill Multifaceted drill ( 8 facets)
Diameter (Ø), mm 6 3/6 6
Flute length, mm 2/15 2/3/15 2/15
Point angle 1400 1300 1180
Lip relief angle 100 100 150
Rack angle 100 100 100
Helix angle 300 300 300
3.3 Selection of Drilling Parameters
The drilling parameters are finalised based on the literature review and some trial runs
taken initially. Spindle speed, feed rate, tool geometry and in particular point angle is
selected as drilling parameters for the experimentation purpose.
FIGURE 3.5 (a) Twist drill
Experimental Work
52
FIGURE 3.5 (b) Step drill
FIGURE 3.5 (c) Multifaceted Drill (8 facets)
3.3.1 Spindle Speed
From the literature survey it can be concluded that to keep the thrust force lower and for
better surface finish drilling operation must be performed at higher cutting speeds. But as
the speed is excessive it wears out the tool which leads to rough surface finish so the
spindle speed selected for this research work is 1500, 2000 and 2500 rpm.
Experimental Work
53
3.3.2 Feed Rate
Feed rate is the speed with which the drill tool is pushed down in to the work material.
Literature review suggests that higher feed rates will lead to increase in thrust force due to
which rough surface finish will be achieved and at the same time if the feed rate is kept
low it will increase the heat generation and lower material removal rates would be
achieved. So the intermediate feed rates are chosen for drilling GFRP as 100, 200 and 300
m/min.
3.3.3 Point Angle
Though the work with different point angle tool is limited it is evident from the literature
review that with increasing point angles the tangential force or the torque at cutting
surfaces decreases and the thrust force increases.so it is better to choose smaller point
angles. Here in the present study three different tool with 1180, 1300 and 1400 point angles
are selected.
Another important parameter for drilling glass fibre reinforced plastics are tool material
and drill diameter. However HSS material is hard enough for cutting GRP and drill
diameter is directly related to the application i.e. requirement of hole size. So HSS as tool
material and 6 mm diameter drills are selected for experimentation purpose.
3.4 Design of Experiment
Taguchi had given an optimization method that is useful in making calculations of
experiments easier and fast. Originally it was designed for the improvement in the quality
of goods. This technique is useful in introducing a method that require a specific set of
experiments to investigate the effectiveness on the response parameters. Orthogonal Array
describes the data structure and the data for the experiments is represented by matrix.
Orthogonal array defines the number of runs during experiment.
MINITAB software is used for Taguchi analysis and the results and graphs were plotted.
The experiments were designed using Taguchi orthogonal array and both the design of
experiment follow the same methodology. The orthogonal array is used to investigate all
Experimental Work
54
the parameters with less number of experiments. Taguchi’s L27 orthogonal array for three
factors, three level experiments, needs 27 runs with 26 degrees of freedom (DoF). In order
to avoid aliasing and overlapping of the interactions of the various factors, only three
columns were chosen from a standard L27 orthogonal array used in the design of
experiments. Table 3.4 shows the L27 orthogonal array and the columns used for the
experimental plan is 1, 2 and 5.
As discussed three parameters are finalised as process parameters i.e. spindle speed, feed
rate and point angle. Three levels of each are selected as shown in Table 3.3. The
responses to be measured in the study are thrust force, torque, delamination at entry and
delamination at exit. These factors are selected based on the thorough literature review.
Each experiment according to Taguchi’s design of experiment was carried out three times
for repeatability and average values are taken for the analysis purpose.
TABLE 3.3 Process parameters and their levels for experiments
Drilling parameters Low level Mid-level High level
Spindle speed v in rpm 1500 2000 2500
Feed rate f in mm/min 100 200 300
Point angle θ in degrees 118 130 140
3.5 Experimental Set Up
Figure 3.6 shows the schematic diagram of experimental set up. Actual set up is shown in
figure 3.7.IITRAM (Institute for Information, Technology, Research and Management)
facilities are used for experiment work. Experiment was carried out on CNC (Computer
Numerical Control) Vertical Machining Centre of Macpower make whose specification are
listed in Table 3.5.
CNC programme is generated using the G an M codes as per the parameters fixed by
Taguchi’s Design of experiment L27 orthogonal array. Piezoelectric dynamometer of
Kistler make Type 9272 is used to measure Thrust force and torque as responses during
study. Specially designed drills with point angles 1180, 1300 and 1400 are changed
accordingly in the drill chuck. The dynamometer comes with four component sensor which
is preloaded between top plate and base plate and it can measure the forces Fx, Fy and Fz
Experimental Work
55
in X, Y and Z direction respectively and it can measure torque around z direction as shown
in the figure of dynamometer. Piezoelectric sensors converts force into electric charge and
TABLE 3.4 Taguchi’s L27 Orthogonal Array
L27
Run 1 2 3 4 5 6 7 8 9 10 11 12 13
1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 1 1 1 1 2 2 2 2 2 2 2 2 2
3 1 1 1 1 3 3 3 3 3 3 3 3 3
4 1 2 2 2 1 1 1 2 2 2 3 2 3
5 1 2 2 2 2 2 2 3 3 3 1 3 1
6 1 2 2 2 3 3 3 1 1 1 2 1 2
7 1 3 3 3 1 1 1 3 3 3 2 3 2
8 1 3 3 3 2 2 2 1 1 1 3 1 3
9 1 3 3 3 3 3 3 2 2 2 1 2 1
10 2 1 2 2 1 2 3 1 2 3 1 3 1
11 2 1 2 2 2 3 1 2 3 1 2 1 2
12 2 1 2 2 3 1 2 3 1 2 3 2 3
13 2 2 3 3 1 2 3 2 3 1 3 1 3
14 2 2 3 3 2 3 1 3 1 2 1 2 1
15 2 2 3 3 3 1 2 1 2 3 2 3 2
16 2 3 1 2 1 2 3 3 1 2 2 2 2
17 2 3 1 2 2 3 1 1 2 3 3 3 3
18 2 3 1 2 3 1 2 2 3 1 1 1 1
19 3 1 3 2 1 3 1 1 3 2 1 2 1
20 3 1 3 2 2 1 2 2 1 3 2 3 2
21 3 1 3 2 3 2 3 3 2 1 3 1 3
22 3 2 1 3 1 3 2 2 1 3 3 3 3
23 3 2 1 3 2 1 3 3 2 1 1 1 1
24 3 2 1 3 3 2 1 1 3 2 2 2 2
25 3 3 2 1 1 3 3 3 2 1 2 1 2
26 3 3 2 1 2 1 1 1 3 2 3 2 3
27 3 3 2 1 3 2 2 2 1 3 1 3 1
Experimental Work
56
FIGURE 3.6 Schematic diagram of experiment set up
FIGURE 3.7 Actual experiment set up
Experimental Work
57
TABLE 3.5 Specification of CNC vertical machining centre (Macpower)
Table size 400 mm x 700 mm
Max. Safe Load on Table 300 Kg
Traverses X axis 510 mm
Y axis 400 mm
Z axis 400 mm
Dist. from Table to Spindle Face 150 - 550 mm
Feed Rates Cutting Feed Rates 1-10000 mm/min.
Rapid Feed Rates X/Y/Z axes 30/30/30 m/min.
Spindle Spindle Taper BT 40
Spindle Speed 80-8000 rpm
Spindle Motor (Cont. Rating) 5.5 kW
Automatic Tool Changer Tool Taper BT 40
Type Twin Arm
Number of Tools 20
Max. Tool Dia. with Adj. 80/125 mm
Max Tool Length 250 mm
Max. Tool Weight 8 kg
Accuracy Positioning 0.010 mm
Repeatability 0.007 mm
controller Siemens 828D
these electric charges are converted in to proportional voltage by a device called Charge
amplifier (Type 5070A) as a voltage signal. Voltage signals are further processed by Data
acquisition system. Data acquisition is the process of measuring an electrical or physical
phenomenon such as voltage, current, temperature, pressure, or sound with the use of
computers. DAQ system consists of sensors, DAQ measurement hardware and a computer
with programmable software. In this study Kistler Dynoware software provided by Kistler
is used which is a universal and easy to use software, which is particularly suitable for
Experimental Work
58
force measurement with dynamometers or single and multi-component force measuring
sensors. Kistler Dynoware software not only measures and stores the data but it is capable
of visualising the curves measured together with graphic functions and calculations. It
gives readings in the form of excel sheet and curves in the MS word. Kistler dynamometer,
Amplifier and DAQ system (Figure 3.10) specifications are shown in table 3.6, 3.7 and 3.8
respectively.
TABLE 3.6 Specification of Kistler Dynamometer Type 9272
Kistler Dynamometer Type 9272
Data Notations Value
Measuring range FX, FY -5 to 5 kN
FZ -5 to 20 kN
MZ -200 to 200 N.m
Overload FX, FY -6 to 6 kN
FZ -6 to 24 kN
M -240 to 240 N.m
Max. bending
moment
MX, M -400 to 400 N.m
Length L 142 mm
Width W 140 mm
Height H 70 mm
Operating
temperature range
0C 0 to 700
Weight 4.2 kg
Sensitivity FX, FY -7 to 8 pC / N
FZ -3 to 5 pC / N
MZ -160 pC / N
Natural frequency Fn (x,y) 3 to 1 KHz
Fn (z) 6 to 3 KHz
Fn (Mz) 4 to 2 KHz
Connector Fischer flange 9-pole negative
Rigidity Cx, Cy 0 to 4 kN/µm
Cz 2 kN/µm
CMz 0 to 7 N.m/µrad
Experimental Work
59
TABLE 3.7 Specification of Multichannel charge Amplifier Type 5070A
Multichannel charge Amplifier Type 5070A
Performance features
4- or 8-Channel Charge Amplifier
6-Component Summing-Calculator (Option)
Measuring range ±200 ... ±200 000 pC or ±600 ...
±600 000 pC
Drift <0.05 pC / s
Liquid crystal display (128x128 Pixel)
Menu-driven user interface
Direct signal evaluation
Adjustment of high-pass and low-pass filters
RS-232C Interface
PC-Software DynoWare
TABLE 3.8 Specification of Data Acquisition System for Force Measurement
Data Acquisition System (DAQ)
Data Unit Value
Dimensions mm 208x70x249
Weight Kg 2.15
Operating
temperature range
°C 0 … 50
Min./max.
temperature
°C –10/60
Input voltage range VDC 10 … 36
Consumption VA 5
Interfaces USB 2.0 (high-
speed)
Number of channels 28
Resolution (per
channel)
Bit 16
Experimental Work
60
FIGURE 3.8 Fixture to hold the laminate
FIGURE 3.9 Dynamometer fitted under fixture
Experimental Work
61
FIGURE 3.10 Charge amplifier and DAQ
FIGURE 3.11 Actual drilling
Experimental Work
62
A fixture (Figure 3.8) is developed to hold the GFRP laminate plate. Fixture is directly
fitted on the dynamometer as shown in Figure 3.9 and laminate plate is drilled with three
tools with different geometry and point angles and each different condition drill is repeated
for three times so total 81 holes were drilled and for measurement of thrust force and
torque average of three is taken. Thrust force and torque has been measured by Kistler
piezo electric dynamometer type 9272.A typical GFRP laminate drilled plate is shown in
figure with coding for reading and measurement purpose. Figure 3.11 shows actual drilling
while Figure 3.12 and 3.13 shows the drilled plate with and without coding.
FIGURE 3.12 Actual drilling without coding
FIGURE 3.13 Actual drilling with coding
Experimental Work
63
3.6 Measurement of Responses
Literature review suggests that to produce good quality holes the responses to be studied
are thrust force, torque and delamination factor.
3.6.1 Thrust force
Thrust force in drilling is the force required to make a hole parallel to axis.it is measured
by Kistler type piezo electric dynamometer in this study. Dynamometer works on the piezo
electric principal which is measured by force sensors. From the literature review it is quite
clear that as the thrust force increases the quality of hole decreases. So to reduce the value
of thrust force it is essential to study the parameters on which thrust force depends. Thrust
force can be studied by analysing various combinations of cutting parameters. For
monitoring, control and documentation generally force is measured as a function of time,
displacement or angle. Dynoware software provided by Kistler is used to plot the force
signals with time. Figure 3.13 shows the plot of thrust force which is recorded as FZ and
drawn by Dynoware software. Moreover software measures FX force in X direction in
Newtons and FY force in Y direction in Newtons and MZ Torque around Z direction in
N.m. Figure 3.14, 3.15 and 3.16 depicts the same. It is found that the thrust force is
maximum when the drill comes in contact with the workpiece and starts drilling and it is
minimum when drill returns.
FIGURE 3.14 Force FX versus Time
Experimental Work
64
FIGURE 3.15 Force FY versus Time
FIGURE 3.16 Force FZ versus Time
FIGURE 3.17 Torque MZ versus Time
Experimental Work
65
3.6.2 Measurement of Torque
Twisting moment required at the external surface of the drill to make a hole is known as
torque and measured in Nm.it can be achieved by product of magnitude of force in
perpendicular plane of axis of rotation and the shortest distance from the axis to the
direction of force. Torque is also the property which effect the hole quality and hence the
detailed study of torque is required to avoid the damages to the drilled hole. Dynoware
software provides the graph of torque versus time which is shown here in Figure 3.17.
3.6.3 Calculation of Delamination factor
While drilling GFRP laminates many defects arises and delamination is one of them.it
happens because of the thrust force of the drill which pushes the layers of plies to
delaminate (or peeling away) rather to cut or drill them. Literature review suggests that
delamination directly varies with the induced thrust force due to increase in wear of the
tool. Delamination also reduces with increase in spindle speed and follows the same
pattern of feed rate. At lower point angles the delamination is observed to be small while it
increases with decrease in point angle it can be because of thrust force reduces at lower
point angles. But it is still demand of time to study the exact effect of the parameters and
the tendency of delamination with the combined effect of these parameters. Delamination
at entry and exit both may have different behaviour. The delamination factor is defined as
the ratio of maximum diameter of the hole including the defects at entry or exit with actual
diameter of drill or hole.
maxd
DF
D (3.1)
Where Fd = delamination factor
Dmax= Maximum diameter of the hole
D = Drill diameter
In other words it can be measured as ratio of Area considering delamination and Area
without delamination.
maxd
AF
A (3.2)
Experimental Work
66
Where Fd = delamination factor
Amax= Maximum Area of the hole considering damage
A = Area of actual hole Drilled
After drilling the holes the measurement was done for delamination at entry and exit. Each
hole is measured under 3D microscope of Mitutoyo make at CHARUSET and with the use
of Image j software as shown in following figure 3.18. Here image j software counts the
area in the form of pixels. Major diameter is considered including the damage area as
shown in figure while minor diameter is measured by considering actual hole area or say
drill dimeter. Delamination factor is obtained by the equation 3.1 for all the 81 readings of
drilled hole.
FIGURE 3.18 Measurement of delamination using IMAGE J software
27 holes with different conditions are required to be drilled as per Taguchi’s design of
experiment and each hole is drilled 3 times. Delamination factor is measured for all the
holes and average is taken for final consideration.
Specification of Mitutoyo Microscope and measurement is shown in Table 3.9 and Figure
3.18 respectively.
Experimental Work
67
TABLE 3.9 Specification of 3D microscope
Make / model Mitutoyo / QS-L2010ZB
Feed mechanism Manual
Observation unit Zoom:0.75 X-5.25X(8X in 7 steps)
Range(X*Y*Z) (200*100*150) mm
Resolution 0.1 µm
Image detecting unit ½ inch colour CMOS camera 3 Mega pixels
Digital zoom 1X-2X-4X
Measuring XY (2.5+20L/1000) µm
Accuracy Z (5+40L/1000) µm
Power consumption 160 W max
FIGURE 3.19 Measurement of drilled hole on 3D Microscope
Experimental Work
68
3.7 Experimental Observations
The experiments are performed as discussed in previous sections and the readings obtained
are summarised in Table 3.10 and 3.11. Each experiment was performed three times for
each set of input as per Taguchi’s and the average values of the responses are tabulated in
order to minimise the error in the experiment.
TABLE 3.10 Observation table for thrust force and delamination at entry Fdentry
Sr
no.
Feed
rate
in
mm
/
min
Cutting
Speed
in rpm
Point
angle in
degree
COD
E
Max.
Hole
area in
mm
2 at
entry
Amax
Hole
area in
mm
2 at
entry A0
Delam
i
nation
factor
Fd=Am
ax/Ao
at
entry
COD
E
Max.
Hole
area in
mm
2 at
entry
Amax
Hole
area in
mm
2 at
entry A0
Delam
i
nation
factor
Fd=Am
ax/Ao
at
entry
COD
E
Max.
Hole
area in
mm
2 at
entry
Amax
Hole
area in
mm
2 at
entry A0
Delam
i
nation
factor
Fd=Am
ax/Ao
at
entry
Thrust
force in
N
Thrust
force in
N
Thrust
force
in N
Average
Thrust
force in
N
Average
Delam
in
ation
factor
Fd=Ama
x/Ao at
entry
1100
1500140
TD1-1
451764252476
1.79TD
2-1363116
2543011.43
TD3-1
428334257008
1.6753.28
66.6537.66
52.531.63
2100
2000140
TD1-1
399806256032
1.56TD
2-1519093
2542922.04
TD3-1
503905250268
2.0170.37
56.2154.14
60.241.87
3100
2500140
TD1-1
417674260124
1.61TD
2-1351006
2493681.41
TD3-1
318120251170
1.2767.93
72.6980.20
73.611.43
4200
1500140
TD4-2
828192256980
3.22TD
5-2574025
2529322.27
TD6-2
457698249827
1.8381.24
63.1158.90
67.752.44
5200
2000140
TD4-2
364041249380
1.46TD
5-2547389
2533922.16
TD6-2
556788254750
2.19102.54
75.0775.26
84.291.94
6200
2500140
TD4-2
346824253866
1.37TD
5-2441220
2476441.78
TD6-2
526096253820
2.07100.83
71.5398.51
90.291.74
7300
1500140
TD7-3
419592254744
1.65TD
8-3428272
2560521.67
TD9-3
786776251600
3.13106.20
103.58118.04
109.272.15
8300
2000140
TD7-3
353600256032
1.38TD
8-3417425
2551761.64
TD9-3
529372223108
2.37113.34
100.04104.74
106.041.80
9300
2500140
TD7-3
384852258340
1.49TD
8-3463247
2524921.83
TD9-3
541060248048
2.1889.11
104.98100.46
98.191.84
10100
1500130
SD1-1
470500256068
1.84SD
2-1423710
2596881.63
SD3-1
362612257008
1.4135.95
61.5257.56
51.681.63
11100
2000130
SD1-1
365866249836
1.46SD
2-1353098
2533921.39
SD3-1
387420256096
1.5155.60
56.6453.96
55.401.46
12100
2500130
SD1-1
319672255152
1.25SD
2-1376148
2542931.48
SD3-1
318741254722
1.2567.87
48.6561.52
59.351.33
13200
1500130
SD4-2
497660251607
1.98SD
5-2627732
2507322.50
SD6-2
512636249824
2.05132.94
49.6885.69
89.442.18
14200
2000130
SD4-2
429470252932
1.70SD
5-2432403
2481041.74
SD6-2
430640245436
1.75119.08
57.4370.56
82.361.73
15200
2500130
SD4-2
544064243196
2.24SD
5-2408193
2471881.65
SD6-2
382612251172
1.52126.40
52.6172.20
83.741.80
16300
1500130
SD7-3
714868247192
2.89SD
8-3497632
2480602.01
SD9-3
507556254744
1.9955.66
92.1664.15
70.662.30
17300
2000130
SD7-3
504380248964
2.03SD
8-3422444
2520501.68
SD9-3
476557250252
1.9064.33
61.2859.02
61.541.87
18300
2500130
SD7-3
604076248084
2.43SD
8-3394212
2520621.56
SD9-3
464958249836
1.8662.50
70.8678.06
70.481.95
19100
1500118
MFD
1-1331226
2534041.31
MFD
2-1312760
2552081.23
MFD
3-1321696
2574461.25
61.7153.28
32.2349.07
1.26
20100
2000118
MFD
1-1292236
2574521.14
MFD
2-1440628
2660401.66
MFD
3-1355636
2502821.42
46.0275.01
55.9759.00
1.40
21100
2500118
MFD
1-1309732
2583181.20
MFD
2-1333911
2542921.31
MFD
3-1326220
2498271.31
52.1255.91
58.5355.52
1.27
22200
1500118
MFD
4-2351964
2484781.42
MFD
5-2411032
2520761.63
MFD
6-2365324
2551961.43
73.6176.54
53.6567.93
1.49
23200
2000118
MFD
4-2326765
2520461.30
MFD
5-2362596
2476381.46
MFD
6-2435905
2516391.73
59.5168.79
64.9464.41
1.50
24200
2500118
MFD
4-2335932
2538641.32
MFD
5-2341040
2431981.40
MFD
6-2328094
2547441.29
58.2338.70
80.5159.14
1.34
25300
1500118
MFD
7-3360472
2538421.42
MFD
8-3433589
2520521.72
MFD
9-3365860
2502821.46
65.8692.22
75.3877.82
1.53
26300
2000118
MFD
7-3364260
2489531.46
MFD
8-3330280
2441101.35
MFD
9-3394262
2458601.60
70.0772.33
76.2972.90
1.47
27300
2500118
MFD
7-3459616
2467021.86
MFD
8-3327240
2520521.30
MFD
9-3412776
2534001.63
55.4285.57
60.0067.00
1.60
Experimental Work
69
TABLE 3.11 Observation table for thrust force and delamination at exit Fdexit
Sr
no.
Feed
rate in
mm
/m
in
Cutting
Speed
in rpm
Point
angle in
degree
CO
DE
Max.
Ho
le
area in
mm
2 at
exit
Am
ax
Ho
le
area. in
mm
2 at
exit A0
Delam
i
nation
factor
Fd=Am
ax/Ao
at exit
CO
DE
Max.
Ho
le
area in
mm
2 at
exit A0
Ho
le
area. in
mm
2 at
exit
Am
ax
Delam
i
nation
factor
Fd=Am
ax/Ao
at exit
CO
DE
Max.
Ho
le
area in
mm
2 at
exit A0
Ho
le
area. in
mm
2 at
exit
Am
ax
Delam
i
nation
factor
Fd=Am
ax/Ao
at exit
Thrust
force
in N
Thrust
force
in N
Thrust
force
in N
Average
Thrust
force in
N
Average
Delam
in
ation
factor
Fd=Am
a
x/Ao
at
exit1
1001500
140TD
1-1430580
2506881.72
TD2-1
509980248944
2.05TD
3-1340554
2524881.35
53.2866.65
37.6652.53
1.70
2100
2000140
TD1-1
347848250294
1.39TD
2-1570000
2471612.31
TD3-1
376666257857
1.4670.37
56.2154.14
60.241.72
3100
2500140
TD1-1
481537248048
1.94TD
2-1324692
2551671.27
TD3-1
320154247188
1.3067.93
72.6980.20
73.611.50
4200
1500140
TD4-2
343646255616
1.34TD
5-2477752
2467481.94
TD6-2
451128257008
1.7681.24
63.1158.90
67.751.68
5200
2000140
TD4-2
526148251194
2.09TD
5-2531948
2533922.10
TD6-2
398096248440
1.60102.54
75.0775.26
84.291.93
6200
2500140
TD4-2
405458252052
1.61TD
5-2358904
2498361.44
TD6-2
369070246259
1.50100.83
71.5398.51
90.291.51
7300
1500140
TD7-3
483376254268
1.90TD
8-3460162
2462791.87
TD9-3
556788243212
2.29106.20
103.58118.04
109.272.02
8300
2000140
TD7-3
464424252954
1.84TD
8-3466106
2591741.80
TD9-3
472928241460
1.96113.34
100.04104.74
106.041.86
9300
2500140
TD7-3
512784252488
2.03TD
8-3449486
2547481.76
TD9-3
378748194776
1.9489.11
104.98100.46
98.191.91
10100
1500130
SD1-1
365343257442
1.42SD
2-1365284
2349841.55
SD3-1
401517257873
1.5635.95
61.5257.56
51.681.51
11100
2000130
SD1-1
374988252492
1.49SD
2-1350452
2485381.41
SD3-1
447680260552
1.7255.60
56.6453.96
55.401.54
12100
2500130
SD1-1
355713258752
1.37SD
2-1316684
2547481.24
SD3-1
417416254748
1.6467.87
48.6561.52
59.351.42
13200
1500130
SD4-2
404880253400
1.60SD
5-2341574
2555941.34
SD6-2
431812258752
1.67132.94
49.6885.69
89.441.53
14200
2000130
SD4-2
433592246288
1.76SD
5-2365866
2587441.41
SD6-2
429984251624
1.71119.08
57.4370.56
82.361.63
15200
2500130
SD4-2
538460256980
2.10SD
5-2355684
2582961.38
SD6-2
369535248508
1.49126.40
52.6172.20
83.741.65
16300
1500130
SD7-3
361024249374
1.45SD
8-3452400
2498361.81
SD9-3
388137254732
1.5255.66
92.1664.15
70.661.59
17300
2000130
SD7-3
530684251640
2.11SD
8-3383192
2556161.50
SD9-3
362636255163
1.4264.33
61.2859.02
61.541.68
18300
2500130
SD7-3
388176245456
1.58SD
8-3363184
2534041.43
SD9-3
326725250280
1.3162.50
70.8678.06
70.481.44
19100
1500118
MFD
1-1344704
2476081.39
MFD
2-1323212
2520761.28
MFD
3-1312760
2551631.23
61.7153.28
32.2349.07
1.30
20100
2000118
MFD
1-1328238
2529481.30
MFD
2-1295570
2516241.17
MFD
3-1329811
2489361.32
46.0275.01
55.9759.00
1.27
21100
2500118
MFD
1-1309784
2560571.21
MFD
2-1306777
2502821.23
MFD
3-1347848
2507131.39
52.1255.91
58.5355.52
1.27
22200
1500118
MFD
4-2343692
2520441.36
MFD
5-2337450
2547461.32
MFD
6-2420848
2569641.64
73.6176.54
53.6567.93
1.44
23200
2000118
MFD
4-2328264
2520761.30
MFD
5-2331316
2467381.34
MFD
6-2349952
2472171.42
59.5168.79
64.9464.41
1.35
24200
2500118
MFD
4-2336876
2507321.34
MFD
5-2323676
2467321.31
MFD
6-2351476
2529341.39
58.2338.70
80.5159.14
1.35
25300
1500118
MFD
7-3344704
2489771.38
MFD
8-3411647
2551761.61
MFD
9-3360458
2533791.42
65.8692.22
75.3877.82
1.47
26300
2000118
MFD
7-3367440
2534001.45
MFD
8-3384852
2516151.53
MFD
9-3336472
2480481.36
70.0772.33
76.2972.90
1.45
27300
2500118
MFD
7-3298496
2502801.19
MFD
8-3331328
2507321.32
MFD
9-3356796
2538661.41
55.4285.57
60.0067.00
1.31
Experimental Work
70
3.8 Summery
This chapter describes the experimental plan, manufacturing of the Glass Fibre Reinforced
Plastic (GFRP) composite Laminate (work piece) and drill tools with their geometry(point
angle ), the Schematic experimental setup and actual set up for drilling holes in the GFRP
Laminate, various responses finalised for analysis along with their measuring techniques
and the instruments and equipment used for the measurements in detail with their
specification and finally the experimental results obtained are summarized. The modelling
and analysis of drilling parameters using the experimental results is presented in the next
chapter.
Results and Discussion
71
CHAPTER – 4
Results and Discussion
For new product design, manufacturing process improvement or process development
experimentation plays a vital role in engineering. The main objective of experiment work
is to develop a robust process, a process which may have least affect by external sources of
variability. The system must be modelled with one or more obtained responses such that
some input gives output to study the inferences. The model is useful for prediction, process
optimisation and control purpose. The modelling technique used here is Adaptive Neuro
Fuzzy Inference System (ANFIS).
4.1 Adaptive Neuro Fuzzy Inference System (ANFIS)
Adaptive Neuro Fuzzy Inference System (ANFIS) is a combination of fuzzy systems and
neural networks. ANFIS is based on hybrid learning methods which can relate an input and
output parameters on the basis of human knowledge in the form of fuzzy if-then rules and
well defined input output pairs. It is a multilayer forward heading network structure which
is having multiple layers of nodes which are interconnected with directional links. All the
nodes are adaptive because as per node parameter it gives the output and based on that it
constructs a learning rule. Learning rule is relation between how to alter input parameters
to lower the given error measure. The basic learning rule of adaptive networks is depend
upon the gradient descent and the chain rule which is slow and has a tendency to become
trapped in local minima. ANFIS is based on hybrid learning rule and it is able to speed up
the learning process.
First order Sugeno fuzzy model is used in this study of ANFIS. i.e. if the fuzzy inference
system has one output ( f ) and three input ( x, y, z) then the first order sugeno model has
following rules:
Rule 1: If x is A1, y is B1 and z is C1 then f1 = p1 x + q1 y + r1z +s1
Results and Discussion
72
Rule 2: If x is A2, y is B2 and z is C2 then f2 = p2 x + q2 y + r2z +s2
Rule 3: If x is A3, y is B3 and z is C3 then f3 = p3 x + q3 y + r3z +s3
For each of the three input variables, x, y and z there are three member ship functions i.e.
(A1, A2, A3), (B1, B2, B3) and (C1, C2, C3). The fuzzy reasoning is illustrated in Figure
4.1, and the corresponding equivalent ANFIS architecture is shown in Figure 4.2.
FIGURE 4.1 Sugeno Fuzzy inference model with three inputs (x, y, z) and one output
(f)
f = 1 1 2 2 3 3 1 2 3( ) ( )w f w f w f w w w
= 1 1 2 2 3 3w f w f w f
Results and Discussion
73
FIGURE 4.2 ANFIS Architecture for three inputs (x, y, z) and one output (f)
Square node is adaptive node while circle node is fixed node. Fixed node do not have
adaptive parameters. Adaptive network is the union of parameter set for each adaptive
node. Parameters are being updated according to the training data and a gradient based
learning procedure to obtain a desired input output mapping. The node functions of each
layer are described below :
Layer 1: Every node, i, is an adaptive node associated with a node function which is
known as membership function. Parameters of membership functions are known as
premise parameters or antecedent parameters.
Grade of membership function is given by
1 ( )i iO A x For i=1, 2, 3…… (4.1)
Where x = input node i,
Ai = linguistic label associated with node function
Results and Discussion
74
1
iO = output of the ith node of layer 1
Generally the bell shaped membership function is selected with highest value equal to 1
and lowest value equal to 0, and is given by
2
1( )
1
i
ii
i
A x
x cb
a
(4.2)
Where ai, bi, ci = premise parameters
The shape of the bell shaped function changes with the change in premise parameters and
creates different types of membership functions according to linguistic label Ai.
Trapezoidal or triangular-shaped membership functions which are continuous and
piecewise differentiable functions are also used as node functions.
Layer 2: In this layer every node is a fixed node labelled which multiplies the incoming
signals and forwards the product out as shown below:
2 ( ) ( ) ( ), 1,2...i i i i iO w A x B y C z i (4.3)
Output of Each node represents the firing strength of a rule.
Layer 3: In this layer also every node is a fixed node labelled N which calculates the ratio
of one rule firing strength to the sum of all rules firing strength. So, the output of this layer
is known as normalized firing strength.
3
1 2 3
, 1,2...iii
wO W i
w w w
(4.4)
Layer 4: In this layer every node is an adaptive node with a node function which is a linear
combination of input variables:
Results and Discussion
75
4ii i i i i i iO W f W p x q y r z s (4.5)
Where iW = output of layer 3
Pi, qi, ri, si = Consequent parameter set
Layer 5: The single node in this layer is a fixed node labelled that computes the
overall output as the summation of all incoming signals,
5 i iiii i i
ii
w ff O W f
w
(4.6)
4.2 Modelling Drilling Parameters Using ANFIS
With the use of ANFIS very accurate model can be achieved relating the drilling
parameters and the responses. MATLAB has ANFIS graphic user interface which is used
in this study for training the ANFIS network. An expert decides the number of rules in a
conventional fuzzy inference system who is involved with the system to be modelled while
in ANFIS, expert is not required because number of membership functions given to each
input parameter is selected empirically. MF examines input and output data and by trial
and error method. This condition is like neural networks. For achieving the desired
performance level minimum number of nodes necessary cannot be decided in advance.
After fixing the number of MFs attached with each input variable, the values of premise
parameters are fixed so that the MFs are correctly spaced along the working range of every
input parameter.
In this study, the input machining parameters are spindle speed, feed rate and point angle
and the output response obtained is thrust force and delamination factor at entry as well as
exit of the hole. The analysis is carried out for the response. Figure 4.3 shows the structure
of a Sugeno type FIS model with three inputs and one output. The output response is thrust
force. Input parameters are speed, federate and point angle. The number of membership
functions for each input is chosen to be three because the error is considerably low for this
choice itself. In MATLAB open Fuzzy Logic Toolbox and then ANFIS editor GUI. This
tool is used for fuzzy inference data modelling.
Results and Discussion
76
FIGURE 4.3 Structure of Sugeno type FIS model with three inputs and one output
Figure 4.4 shows the structure of a typical ANFIS model used in the present study. This
figure shows the layer 1 to layer 5 interconnections in ANFIS training model. The number
of rules generated are 27.
FIGURE 4.4 ANFIS Structure for three input and one output
Then 27 training data sets are considered as input training vectors and the function ANFIS
is used to train the FIS model. The type of membership functions are chosen by trial and
error and hence four types of membership functions namely, generalized bell shaped MF,
triangular MF, Gaussian MF and two-bell Gaussian MF are applied in this study. Here, the
Results and Discussion
77
parameter optimization method is chosen to be ‘Hybrid’ and the number of epochs for
ANFIS training is set to 40. The training stops after the training data error remains within
the error tolerance that is chosen as zero or if the training epoch number is reached. ANFIS
information for various types of membership functions are tabulated in Table 4.1. It is
evident that the number of nonlinear parameters differs for Gaussian functions.
TABLE 4.1 ANFIS information for different MFs
Sr.no ANFIS
parameters
Type of MF
Triangular
MF(trimf)
Generalised
Bell shaped
MF (gbellmf)
Gaussian
MF
(gaussmf)
Two
Gaussian MF
(gauss2mf)
1 No. of Nodes 78 78 78 78
2 No. of linear
parameters
27 27 27 27
3 No. of non-
linear
parameters
27 27 18 36
4 Total No. of
parameters
54 54 45 63
5 No. of fuzzy
rules
27 27 27 27
Figure 4.5 shows the 27 rules for given three input parameters and one output response.
Same procedure is followed for other two output responses, i.e. delamination at exit and
delamination at entry. These rules are used to predict the output responses i.e. thrust force,
delamination at entry and delamination at exit by combination of any values of the input
parameters randomly within the given range.
4.3 Effect of Input Variables in Drilling of GFRP
In this study drilling experiment is carried out at three levels (low level, mid-level and high
level) of drilling parameters or input variables. Drills used are having geometry of twist
drill with point angle of 1400, step drill of 1300 and multifaceted drill with point angle of
1180 and all are of 6 mm diameter. The responses obtained are thrust force, torque,
delamination at entry and delamination at exit. All these responses have “lower the better”
qualities. A best quality hole is only possible with lowest values of each responses. But it is
next to impossible to obtain the lowest values of each responses every time so the optimum
Results and Discussion
78
drilling conditions are identified at which the combination of low values for each responses
is achievable.
FIGURE 4.5 Rules for three input and one output
4.3.1 Effect of Drilling Parameters on Thrust Force
As per literature review thrust force is the main responsible parameter for delamination and
its lower value will result in good quality hole with less delamination. For a specific
drilling conditions thrust force is recorded using Kistler dynamometer which is shown
graphically in Figure 4.6. Similarly the values are obtained for all the drilling conditions
with three types of drills and are tabulated in Table 3.10 and 3.11.
Results and Discussion
79
Twist drill with point angle 1400
Step drill with point angle of 1300
Multi-faceted drill with point angle 1180
FIGURE 4.6 Thrust force obtained at 100 m/min feed rate and speed of 1500, 2000
and 2500 rpm
Results and Discussion
80
4.3.2 Effect of Drilling Parameters on Torque
Twist drill with point angle 1400
Step drill with point angle of 1300
Multi-faceted drill with point angle 1180
FIGURE 4.7 Torque at 100 m/min feed rate and speed of 1500, 2000 and 2500 rpm
Results and Discussion
81
Torque is the moment around the Z direction. As per literature review Torque is the also
responsible parameter for delamination and its lower value will result in good quality hole
with less delamination. For a specific drilling conditions torque is recorded using Kistler
dynamometer which is shown graphically in Figure 4.7. Similarly the values are obtained
for all the drilling conditions with three types of drills and are tabulated in Table 3.10 and
3.11.
4.4 ANOVA Analysis of Drilling Parameters and Plots Showing the
Interaction Effect of Drilling Parameters in Drilling of GFRP Composites
Analysis of variance (ANOVA) is statistical tool for data analysis. It is a collection of
statistical models, and their associated procedures, in which the observed variance is
partitioned into components due to different explanatory variables. ANOVA is useful to
find the parameters which are individually or in combination significantly affect the
drilling process. The effect of drilling parameters were studied in earlier chapters so to find
the effect of interactions, the most affecting factor and deviation from actual values
statistical analysis like ANOVA is performed in the present study.
Statistical method ANOVA (Analysis of Variance) was performed for Thrust force,
Delamination at entry and Delamination at exit of GFRP Laminate for checking the
significance level of each parameter (input Variables) and it is also used to find the
percentage contribution of Point Angle, Feed Rate and Cutting Speed at entry and at exit.
ANOVA is done for 95% confidence level so if p-value is less than 0.05 then that factor is
responsible factor for the output.
4.4.1 Analysis of Thrust Force using ANOVA
TABLE 4.2 ANOVA table for Thrust force F
Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value
Feed rate in mm/min 2 2933.24 1466.62 12.22 0
Cutting Speed in rpm 2 24.89 12.45 0.1 0.902
point angle(tool geometry) 2 1674.51 837.25 6.98 0.005
Error 20 2400.45 120.02
Total 26 7033.1
S R-sq R-sq(adj) R-sq(pred)
10.9555 85.87% 55.63% 37.80%
Results and Discussion
82
ANOVA result for thrust force is tabulated in Table 4.2 from all the three parameters i.e.
Spindle speed, feed rate and point angle, feed rate is the most influential factor because its
F value (12.22) is higher in compare to cutting speed and point angle. Hence, it is found
that the feed rate has a more dominant effect on Thrust force than point angle and spindle
speed. The next contributing factor is point angle which is then followed by spindle speed
For Thrust force the factors feed rate and point angle are significant (p<0.05) whereas
spindle speed is insignificant (p>0.05).
The value of the coefficient of determination (R²) indicates that 85.87% of the variability
in the response could be explained by the model.
The perfect behaviour of the response with change in one factor is controlled by the
constant values of other variables. So it is important to investigate the combined effects of
the drilling parameters.
MATLAB has a facility of surface viewer to plot 3D surfaces. In the present study sugeno
fuzzy model in ANFIS is tested and nonlinear surfaces for the problem of drilling GFRP
laminates are plotted. The surface viewer is a facility in MATLAB where one can check
the interaction of input parameters to its responses using ANFIS model.3-D plots in the
surface viewer depicts the fuzzy surface of the trained approximator.
(a) At feed rate of 100 m/min (b) At feed rate of 200 m/min
(c) At feed rate of 300 m/min (d) At spindle speed of 1500 rpm
Results and Discussion
83
(e) At spindle speed of 2000 rpm (f) At spindle speed of 2500 rpm
(g) At point angle of 1180 (h) At point angle of 1300
(i) At point angle of 1400
FIGURE 4.8 Effect of spindle speed, point angle and feed rate on thrust force
Figure 4.8 shows the surface plots achieved using ANFIS model which correlates the
interaction effects of spindle speed, feed rate and point angle on thrust force in drilling of
GFRP laminates. Figure 4.8(a) shows the effect of speed and point angle on thrust force at
feed rate of lower value i.e. 100 m/min. It is investigated that increase in spindle speed
initially increases thrust force but with more increase in spindle speed it almost remains
constant. At the same time with increase in point angle and spindle speed thrust force
increases drastically. This can be explained as compared to a broader point angle, the
narrow point angle has less contact area between the GFRP laminate and the drill hence
lower is the thrust force required to drill. However it is evident from the plot that the effect
of spindle speed is very less as compared to the point angle on thrust force.
Results and Discussion
84
It is observed that
A linear co relation of thrust force exists at smaller point angles
The graph of thrust force increases as point angle increases and becomes nonlinear
from its mid value onwards.
Same behaviour can be seen (refer Figure 4.8(b) and Figure 4.8(c)) at higher levels of feed
rate i.e. 300 m/min but at mid-levels an increase in point angles from 1180 to 1300
increases the thrust force drastically this is because the 1300 point angle drill is step drill
which has the stepping effect so when 6 mm diameter inserts in to laminate it increases
thrust force abnormally.
Figure 4.8(d) depicts the interaction effect of point angle and feed rate on thrust force as
these two are the main factors responsible for increase in thrust force. It can be seen from
the plot that only increase in point angle increases thrust force linearly but combined
increase in feed rate and point angle increases thrust force sharply which indicates that the
feed rate is the main influential factor in this plot. Highest variation in thrust force is
noticed at the higher values of feed rate. This behaviour was observed at spindle speed of
1500 rpm but the same can be studied from the plots of spindle speed at 2000 rpm and
2500 rpm (Refer Figure 4.8(e) and Figure 4.8 (f)).
Surfaces Plotted in Figure 4.8(g), Figure 4.8(h) and Figure 4.8(i) are showing the
interaction effect between cutting speed and feed rate on thrust force at point angle of 1180,
1300 and 1400 respectively. Very small increase in thrust force noticed at lower spindle
speed to higher spindle speeds at all point angles but with increase in feed rates from lower
values to higher values thrust force increases sharply and linearly. Abrupt increase in thrust
force can be seen at mid values of feed rates for point angle of 1300 because of tool
geometry which is step drill geometry. It is investigated that geometry of drill in addition
to point angle effects thrust force.
4.4.2 Analysis of Delamination at Entry (Fdi) using ANOVA
ANOVA result for Delamination at Entry (Fdi) is tabulated in Table 4.3 from all the three
parameters i.e. Spindle speed, feed rate and point angle, point angle is the most influential
Results and Discussion
85
factor because its F value (17.36) is higher in compare to cutting speed and point angle.
Next influential factor is feed rate having F value 11.96 followed by cutting speed.
For Delamination Factor at Entry the factors Point Angle, Feed Rate and Cutting Speed all
are Significant (p<0.05). Hence, it is found that the Point Angle, Feed Rate and Cutting
Speed all have dominant effect on Delamination Factor at Entry.
The value of the coefficient of determination (R²) indicates that 77.59% of the variability
in the response could be explained by the model.
TABLE 4.3 ANOVA table for Delamination at Entry Fdi
(a) At feed rate of 100 m/min (b) At feed rate of 200 m/min
(c) At feed rate of 300 m/min (d) At spindle speed of 1500 rpm
Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value
Feed rate in mm/min 2 0.6983 0.34914 11.96 0
Cutting Speed in rpm 2 0.3099 0.15496 5.31 0.014
point angle(tool geometry) 2 1.0132 0.50661 17.36 0
Error 20 0.5837 0.02918
Total 26 2.6051
S R-sq R-sq(adj) R-sq(pred)
0.170834 77.59% 70.87% 59.17%
Results and Discussion
86
(e) At spindle speed of 2000 rpm (f) At spindle speed of 2500 rpm
(g) At point angle of 1180 (h) At point angle of 1300
(i) At point angle of 1400
FIGURE 4.9 Effect of spindle speed, point angle and feed rate on Delamination at
Entry
Figure 4.9 shows the surface plots achieved using ANFIS model which correlates the
interaction effects of spindle speed, feed rate and point angle on thrust force in drilling of
GFRP laminates. Figure 4.8(a) shows the effect of speed and point angle on thrust force at
feed rate of lower value i.e. 100 m/min. Plot shows that as point angle increases,
delamination factor at entry (Fdi) increases significantly whereas with increase in spindle
speed it decreases. From Figure 4.8(b) and Figure 4.8(c) it can be concluded that at higher
speeds with higher point angle also the entry delamination factor can be controlled within
given range. It is also evident from the plots 4.8(a), 4.8(b) and 4.8(c) that at lower feed
rates (100 m/min) entry delamination factor is minimum.
Figure 4.8(d), 4.8(e) and 4.8 (f) shows the interaction effect of point angle and feed rate on
Delamination factor at Entry at spindle speed of low, medium and high value. It can be
Results and Discussion
87
seen from the plots that with increase in point angle and feed rates entry delamination
factor increases almost constantly. It is also evident that the combination of lower feed
rates and higher spindle speeds has minimum entry delamination factor even with higher
point angle tool geometry.
Surfaces Plotted in Figure 4.8(g), Figure 4.8(h) and Figure 4.8(i) are showing the
interaction effect between spindle speed and feed rate on entry delamination factor at point
angle of 1180, 1300 and 1400 respectively. It is very clear from the plots that as the feed
rates increases the entry delamination factor increases. So the feed rate is main influential
factor in causing damage at entry in drilling of GFRP laminates. The main aim of the
present study is to monitor and control the damage and delamination caused during drilling
of GFRP laminates. From the above said results, it is concluded that delamination can be
reduced at lower feed rates and with lower point angle drill tools.
4.4.3 Analysis of Delamination at Exit (Fdo) using ANOVA
ANOVA result for Delamination at Exit (Fdo) is tabulated in Table 4.4 from all the three
parameters i.e. Spindle speed, feed rate and point angle, point angle is the most influential
factor because its F value (39.2) is higher in compare to cutting speed and point angle.
Followed by feed rate (F value 6.69) and cutting speed (F value 3.77).
TABLE 4.4 ANOVA table for Delamination Factor at Exit Fdo
Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value
Feed rate in mm/min 2 0.12564 0.062818 6.69 0.006
Cutting Speed in rpm 2 0.07085 0.035426 3.77 0.041
point angle(tool geometry) 2 0.73628 0.368141 39.2 0
Error 20 0.18782 0.009391
Total 26 1.12059
S R-sq R-sq(adj) R-sq(pred)
0.0969075 83.24% 78.21% 69.45%
For Delamination Factor at Exit the factors Point Angle, Feed Rate and Cutting Speed all
are significant (p<0.05). Hence, it is found that the Point Angle, Feed Rate and Cutting
Speed all have dominant effect on Delamination Factor at Exit. The value of the coefficient
of determination (R²) indicates that 83.24% of the variability in the response could be
explained by the model.
Results and Discussion
88
(a) At feed rate of 100 m/min (b) At feed rate of 200 m/min
(c) At feed rate of 300 m/min (d) At spindle speed of 1500 rpm
(e) At spindle speed of 2000 rpm (f) At spindle speed of 2500 rpm
(g) At point angle of 1180 (h) At point angle of 1300
(i) At point angle of 1400
FIGURE 4.10 Effect of spindle speed, point angle and feed rate on Delamination at
Exit
Results and Discussion
89
Figure 4.10 shows the surface plots achieved using ANFIS model which correlates the
interaction effects of spindle speed, feed rate and point angle on Delamination Factor at
Exit in drilling of GFRP laminates. Figure 4.9 (a) shows the effect of speed and point angle
on Delamination Factor at Exit at feed rate of lower value i.e. 100 m/min. Plot shows that
as point angle increases, delamination factor at exit (Fdo) of the hole increases significantly
whereas with increase in spindle speed it decreases. At lower point angles and higher
spindle speeds delamination at exit can be avoided. From Figure 4.9 (b) and Figure 4.9 (c)
it can be learnt that point angle and delamination at exit have direct relationship. Figure 4.9
(d), 4.9 (e) and 4.9 (f) shows the interaction effect of point angle and feed rate on
Delamination factor at Exit at spindle speed of low, medium and high value. Plots suggests
that with increase in point angle and feed rates exit delamination factor increases linearly.
It is also evident that the combination of lower feed rates and higher spindle speeds has
minimum exit delamination factor even with higher point angle tool geometry. Figure 4.8
(g), Figure 4.8(h) and Figure 4.8 (i) are showing the interaction effect between spindle
speed and feed rate on exit delamination factor at point angle of 1180, 1300 and 1400
respectively. As the feed rate increases the exit delamination factor also increases. At
lower point angles and higher spindle speeds delamination factor at exit of the hole can be
controlled.
The summary of interaction studies analysed using the ANOVA tables and the above
surface plots suggests that feed rate is the most influential factor for drilling GFRP
laminates while the influential interaction may vary for every response.
From all above analysis the conclusion is that a good combination of low point angle, low
feed rate and high spindle speed can result in
(1) Lower thrust forces which leads to less damage in drilling of GFRP
laminates.
(2) Lower delamination factor at entry (Fdi) in drilling of GFRP
(3) Lower delamination factor at exit ( Fdo) in drilling of GFRP
Thus the surface viewer of MATLAB is used for interaction plots with clear 3D surfaces
for prediction of various responses for any given input variables of drilling or to obtain an
optimum drilling condition for minimum damage in drilling of GFRP composite plates.
Experimented results are in close approximation to the predicted values with more than
Results and Discussion
90
92% accuracy. Above results shows that ANFIS model can predict the thrust force and so
based on the model cutting parameters can be selected. Comparison of predicted and
experimented values are shown in Table 4.5.
TABLE 4.5 Comparison of predicted and experimented values
Sr
no. Point Angle
Feed rate in
mm/min
Cutting
Speed in
rpm
Actual
Average
Thrust
force in N
Predicted
by ANFIS
model
%
accuracy
1 118 150 1700 67.75 59.40 87.68
2 118 225 2100 59.88 65.20 91.11
3 118 280 2375 65.49 67.00 97.70
4.5 Summery
A detailed analysis of drilling mechanism is discussed and the reason for delamination is
identified. The images taken by 3D microscopes are illustrated and delamination at entry
and exit are clearly shown. Interaction plots between spindle speeds, point angles and feed
rates are discussed in detail and ANOVA analysis for various factors have been studied to
identify the responsible factor for causing thrust force and damages.it is investigated that
the feed rates are the most influential factor in drilling of GFRP and optimum drilling
conditions can be achieved with the careful selection of all the three input parameters.
Conclusions and Future Scope
91
CHAPTER – 5
Conclusions and Future Scope
5.1 Conclusions
GFRP composites are widely used in high performance engineering applications like
automotive parts and aerospace industry, health services and sports and energy
applications. By the selection of appropriate fibre, its orientation and matrix material, it is
possible to fabricate composite laminates with desired properties. Drilling and other
machining operations are essential for converting laminates in to required shapes though
laminates are fabricated near to net shape. The challenges of tool wear and damages
produced in machining restricts the application of GFRP in many fields. It is the demand of
time to optimise the machining process of composites to make them eligible for specific
applications. The factors affecting the optimisation of GFRP machining, specifically
drilling, is addressed in this research work. The main objective of this research work is to
monitor the effect of cutting parameters and tool geometry on thrust force and torque
which are the responsible factors for creating delamination at entry as well as exit and to
develop a mathematical model which can be a readymade tool for the selection of cutting
parameters and point angles of tool for delamination free drilling.
The present research is contributing in knowledge of GFRP drilling using tools having
point angles of 1180, 1300 and 1400. Based on Taguchi’s L27 orthogonal array the
experiments were designed and conducted. Cutting parameters selected were feed rate,
spindle speed and point angle and value of each are selected at low, medium and high
level. CNC machining centre is used to perform the experiment. Kistler Dynamometer was
used to measure the thrust force and torque. Mitutoyo 3D microscope was used to measure
delamination factor at entry (Fdi) and delamination factor at exit (Fdo). Optimum values
cutting parameters and point angles are identified for delamination free drilling within
given range of cutting parameters.
Conclusions and Future Scope
92
Mathematical model is developed in MATLAB using ANFIS (Adaptive Neuro Fuzzy
Inference System) tool which gives combine advantage of fuzzy logic theories and neural
network system concept to predict the optimum values for delamination free drilling within
selected range of cutting parameters. Confirmation experiments were performed to check
the predicted values and the results are found in close approximation. The interconnection
of cutting parameters and responses are discussed with the help of ANFIS surface plots and
ANOVA analysis. Following points are concluded from the research work:
1. ANOVA analysis and ANFIS surface plots suggests that most responsible factor for
drilling is feed rate in GFRP laminates while the influential interaction may vary for
every response obtained. Delamination factor at entry (Fdi) and exit (Fdo) is directly
varies with Feed rate in drilling of GFRP laminates. Hence low feed rate of 100
m/min is best preferred in drilling of GFRP.
2. Torque, Thrust force, Delamination factor at entry (Fdi) and exit (Fdo) decrease with
increase in the spindle speed. So it is well judge that drilling operation must be
carried out at higher speeds and in this case 2500 rpm is well suited for drilling in
GFRP laminates.
3. It is concluded that all the response factors increase as the value of point angle
increases. Hence the lowest value of point angle which is 1180 and having
multifaceted tool geometry is best for drilling in GFRP laminates.
4. Statistical analysis using ANOVA technique is done to find out the most influencing
factor out of all cutting parameters and most responsible factor concluded is feed
rate and point angle is the second responsible factor followed by spindle speed.
5. A model is developed in MATLAB software using ANFIS. Using this model the
values of cutting parameters are predicted for delamination free drilling and
experimentally it is found that predicted values by ANFIS model are in close
approximation with the experimented values. Model can predict values of cutting
parameters for delamination free drilling with more than 92 % accuracy.
6. Multifaceted drill with 118 degree point angle, 100 mm/min feed rate and 1500 rpm
have minimum delamination at entry (1.26) and exit (1.3). Minimum value of thrust
force obtained was 49.07 N.
Conclusions and Future Scope
93
7. Most preferred drilling conditions observed at lowest point angle and lowest feed
rate and highest spindle speed within selected range of cutting parameters.
5.2 Future scope
1. In this research work the laminate sheet used was of 4 mm thickness. There are
certain applications where higher thickness sheets are used so GFRP plates with
different thicknesses can be taken and effect of thickness can be studied.
2. Prediction accuracy can be increased by selecting more than three number of levels
for cutting parameters in the selected ranges and by performing more practical if
cost and time is not a barrier.
3. Types of tools with different geometry and materials can be studied for effect on
variables.
4. Different statistical methods and data analysis method can be used.
5. By taking in consideration all the given points in conclusion few more experiments
can be performed like cover plate while drilling, pack drilling or waterjet drilling
for more insight.
6. Self-healing materials (carbon Nano tubes filled with bonding material) can be
investigated to reduce delamination.
References
94
List of References
[1] Zhang LB, Wang LJ, Wang X. (2003). Study on vibration drilling of fiber reinforced
plastics with hybrid variation parameters method. Composites: Part A 2003; 34:237–
44.
[2] Davim, J. P (2001). “A note on the determination of optimal cutting conditions for
surface finish obtained in turning using design of experiments”, J. Mater. Proces.
Technol., Vol. 116, pp. 305-308
[3] Hussain, S. A., Pandurangadu, V., & Kumar, K. P. (2011). Machinability of glass fiber
reinforced plastic (GFRP) composite materials. International Journal of Engineering,
Science and Technology, 3(4), 103–118.
[4] Ali, H. M., Iqbal, A., & Liang, L. I. (2013). A comparative study on the use of drilling
and milling processes in hole making of GFRP composite, 38(August), 743–760.
[5] Hu, N.S. and Zhang, L.C. (2004) “Some observations in grinding unidirectional carbon
fibre-reinforced plastics”, J. Mater. Proces. Technol., Vol. 152, pp. 333-338.
[6] Lachaud, F., Piquet, R., Collombet, F., & Surcin, L. (2001). Drilling of composite
structures. Composite Structures, 52(3–4), 511–516.
[7] Hocheng, H. and Dharan, C. K. H. (1990). “Delamination during drilling of composite
laminates”, Transactions of the ASME, J. Engg. Industry, Vol. 112, pp. 236-239.
[8] Zhang, H., Chen, W., Chen D. and Zhang, L. (2001). “Assessment of the exit defects in
carbon fibre-reinforced plastic plates caused by drilling”, Trans. Tech. Publications,
Switzerland, Key Engineering Materials, Vol. 196, pp. 43-52.
[9] DiPaolo, G., Kapoor, S.G. and DeVor, R.E. (1996). “An experimental Investigation of
the Crack Growth Phenomenon for Drilling of fiberreinforced Composite Materials,
Transactions of the ASME”, J. Engg. Industry, Vol. 118.
[10] Mathew, J., Ramakrishnan, N. and Naik, N. K. (1999) “Trepanning on Unidirectional
composites”, Composites, Part A, Vol. 30, pp. 951-959.
[11] Piquet, R., Ferret, B., Lachaud, F. and Swider, P. (2000). “Experimental analysis of
drilling in thin carbon/epoxy plate using special drills”, Composites: Part A, Vol. 31,
pp. 1107-1115.
[12] Chen, W. C. (1997), “Some experimental investigations in the drilling of carbon fiber-
reinforced plastic (CFRP) composite laminates”, Int. J. Mach. Tools Manuf. Vol. 37,
No. 8, pp. 1097-1108.
References
95
[13] Hocheng, H. and Tsao, C. C. (2006). “Effects of special drill bits on drilling induced
delamination of composite materials”, Int. J. Mach. Tools Manufact. Vol.46, pp.1403-
1416.
[14] Zitoune, R., & Collombet, F. (2007). Numerical prediction of the thrust force
responsible of delamination during the drilling of the long-fibre composite structures.
Composites Part A: Applied Science and Manufacturing, 38(3), 858–866.
[15] Davim, J. P., Rubio, J. C., & Abrao, A. M. (2007). A novel approach based on digital
image analysis to evaluate the delamination factor after drilling composite laminates.
Composites Science and Technology, 67(9), 1939–1945.
[16] Palanikumar, K., Rubio, J. C., Abrao, A., Esteves, A., & Davim, J. P. (2008).
Statistical Analysis of Delamination in Drilling Glass Fiber-Reinforced Plastics
(GFRP). Journal of Reinforced Plastics and Composites, 27(15), 1615–1623.
[17] Cenna, A.A. and Mathew, P. (2002). “Analysis and prediction of laser cutting
parameters of fibre reinforced plastics (FRP) composite materials”, Inter. J. Mach.
Tools Manuf., Vol. 42, pp. 105-113.
[18] Liakus, J., Wang, B., Cipra, R. and Siegmund, T. (2003) “Processingmicrostructure-
property predictions for short fiber reinforced composite structures based on a spray
deposition process”, Composite Struct., Vol. 61, pp. 363-374.
[19] Gordon, S. and Hillery, M. T. (2003) “A review of the cutting of composite
materials”, Proc. Instn. Mech. Engrs., Part I: J. Mater.: Desgn and applications, Vol.
217, pp. 35-44.
[20] Davim, J. P. and Reis, P. (2005) “Damage and dimensional precision on milling
carbon fiber-reinforced plastics using design experiments”, J. Mater. Proces.
Technol’, Vol. 160, pp. 160-167.
[21] Davim, J. P. and Reis, P. (2003). “Study of delamination of in drilling carbon fiber
reinforced plastics (CFRP) using design experiments”, Composite Structures, Vol. 59,
pp. 481-487.
[22] Linbo, Z., Lijiang, W., & Xin, W. (2003). Study on vibration drilling of fiber
reinforced plastics with hybrid variation parameters method. Composites Part A:
Applied Science and Manufacturing, 34(3), 237–244.
[23] Edoardo Capello (2004). Workpiece damping and its effect on delamination damage
in drilling thin composite laminates. Journal of Materials Processing
Technology 148(2):186-195
[24] Khashaba, U. A. (2004). Delamination in drilling GFR-thermoset composites.
Composite Structures, 63(3–4), 313–327.
[25] Mohan, N. S., Ramachandra, A., & Kulkarni, S. M. (2005). Machining of Fibre-
reinforced Thermoplastics: Influence of Feed and Drill Size on Thrust Force and
References
96
Torque during Drilling. Journal of Reinforced Plastics and Composites, 24(12),
1247–1257.
[26] Tsao, C. C. (2008). Influence of drill geometry in drilling carbon fibre reinforced
plastics. In Key Engineering Materials (Vol. 375–376, pp. 236–240).
[27] Oliver and Ekkard (2014). “Tool wear analyses in low frequency vibration assisted
drilling of CFRP/Ti6Al4V stack material” 6th CIRP International Conference on High
Performance Cutting, HPC2014, 142 – 147.
[28] Dharan, C.K.H. and Won, M.S., (2000). “Machining parameters for an intelligent
machining system for composite laminates”, Inter. J. Mach. Tools Manuf., Vol. 40,
pp. 415-426.
[30] Velayudham, A., & Krishnamurthy, R. (2007). Effect of point geometry and their
influence on thrust and delamination in drilling of polymeric composites. Journal of
Materials Processing Technology, 185(1–3), 204–209.
[31] Latha, B., Senthilkumar, V. S., & Palanikumar, K. (2011). Influence of drill geometry
on thrust force in drilling GFRP composites Journal of Reinforced Plastics and
Composites. Journal of Reinforced Plastics and Composites.
[32] Satsangi, P. S. (2012). A genetic algorithmic approach for optimization of surface
roughness prediction model in turning using UD-GFRP composite, 19(December),
386–396.
[33] Grilo, T. J., Paulo, R. M. F., Silva, C. R. M., & Davim, J. P. (2013). Experimental
delamination analyses of CFRPs using different drill geometries. Composites Part B:
Engineering. 1344–1350.
[34] Karpat, Y., & Bahtiyar, O. (2015). Comparative analysis of PCD drill designs during
drilling of CFRP laminates. Procedia CIRP, 31, 316–321.
[35] Durao, L. M. P., Goncalves, D. J. S, Tavares, J. M. R. S., de Albuquerque, V. H. C.
and Marques, A. T. (2010). “Drilling process of composite laminates-A tool based
analysis”, 14th, European Conference on Composite Materials, pp, 1-10.
[36] Palanikumar, K. (2011). Experimental investigation and optimisation in drilling of
GFRP composites. Measurement, 44(10), 2138–2148.
[37] Lin, S. C. and Chen, I. K. (1996). “Drilling carbon fiber-reinforced composite material
at high speed”, Wear, Vol. 194, pp. 156-162.
[38] Fernandes and Cook (2006). “Drilling of carbon composites using a one shot drill bit.
Part II: Empirical modelling of maximum thrust force” International Journal of
Machine Tools and Manufacture 46(1):76-79
[39] Tsao, C. C. and Hocheng, H. (2007). “Parametric study on thrust force of core drill”,
J. Mater. Proces. Technol., Vol. 192-193, pp. 37-40.
References
97
[40] Jayabal, S., & Natarajan, U. (2010). Influence of cutting parameters on thrust force
and torque in drilling of E-glass/polyester composites. Indian Journal of Engineering
and Materials Sciences, 17(6), 463–470.
[41] Rahamathullah, I., & Shunmugam, M. S. (2011). Thrust and torque analyses for
different strategies adapted in microdrilling of glass-fibre-reinforced plastic.
Proceedings of the Institution of Mechanical Engineers, Part B : Journal of
Engineering Manufacture, 225(4), 505–519.
[42] Vankanti and Ganta (2013). “Optimization of process parameters in drilling of GFRP
composite using Taguchi method”, Journal of Materials Research and Technology
Volume 3, Issue 1, Pages 35-41
[43] Su, Wang, Yuan, and Cheng (2015). “Study of thrust forces and delamination in
drilling carbon-reinforced plastics (CFRPs) using a tapered drill-reamer”, The
International Journal of Advanced Manufacturing Technology Volume 80, Issue 5–
8, pp 1457–1469|
[44] Tagliaferri, V., Caprino, G. and Diterlizzi, A. (1990). “Effect of drilling parameters on
the finish and mechanical properties of GFRP Composites”, Int. J. Mach. Tools
Manuf., Vol. 30, No. 1, pp. 77-84.
[45] Chen, W. C., (1997). “Some experimental investigations in the drilling of carbon
fiber-reinforced plastic (CFRP) composite laminates”, Int. J. Mach. Tools Manuf.
Vol. 37, No. 8, pp. 1097-1108.
[46] Hocheng, H. and Tsao, C. C. “Effects of special drill bits on drillinginduced
delamination of composite materials”, Int. J. Mach. Tools Manufact. Vol.46, pp.1403-
1416, 2006.
[47] Gaitonde, V. N., Karnik, S. R., & Paulo Davim, J. (2008). Taguchi multiple-
performance characteristics optimization in drilling of medium density fibreboard
(MDF) to minimize delamination using utility concept. Journal of Materials
Processing Technology, 196(1–3), 73–78.
[48] Fotouhi, M., Pashmforoush, F., Ahmadi, M., & Refahi Oskouei, a. (2011). Monitoring
the initiation and growth of delamination in composite materials using acoustic
emission under quasi-static three-point bending test. Journal of Reinforced Plastics
and Composites, 30(17), 1481–1493.
[49] Kilickap, E. (2011). Analysis and modeling of delamination factor in drilling glass
fiber reinforced plastic using response surface methodology. Journal of Composite
Materials, 45, 727–736.
[50] Khaled Giasin, Sabino Ayvar and A. Hodzic (2016), “An experimental study on
drilling of unidirectional GLARE fibre metal laminates, Composite Structures 133.
References
98
[51] Konig, W., Wulf, Ch., Gra, P. and Willerscheid, H. (1985). “Machining of Fibre
Reinforced Plastics”, Ann. CIRP, Vol. 34, No. 2, pp. 537-548.
[52] Caprino, G. and Tagliaferri, V. (1995). “Damage development in drilling glass fibre
reinforced plastics”, Int. J. Mach. Tools Manuf., Vol. 35, No. 6, pp. 817-829, 1995.
[53] Langella, A., Nele, L. and Maio, L. (2005). “A torque and thrust prediction model for
drilling of composite materials”, Composites: Part A Appd. Sci. Manuf., Vol. 36, pp.
83-93.
[54] Tsao, C. C. (2006). “Taguchi analysis of drilling quality associated with core drill in
drilling of composite material”, Int. J. Adv. Manuf. Technol., Vol. 32, Nos. 9-10, pp.
877-884.
[55] Krishnaraj, V. (2008). Effects of drill points on glass fibre reinforced plastic
composite while drilling at high spindle speed. Proceedings of the World Congress on
Engineering, 2, 2–4.
[56] Durao, L. M. P., Goncalves, D. J. S, Manuel, J., Tavares, J. M. R. S., de Albuquerque,
V. H. C., Vieira, A. A. and Marques, A. T. (2010). “Drilling tool geometry evaluation
for reinforced composite laminates”, Composite Structures, Vol. 92, No. 7, pp. 1545-
1550.
[57] Wang, D. H., Ramulu, M. and Arola, D. (1995). “Orthogonal cutting mechanisms of
graphite/epoxy composite. Part II: Multi-directional laminate”, Int. J. Mach. Manuf.,
Vol. 35, No. 12, pp. 1639-1648.
[58] George, P. M., Raghunath, B., Manocha, L. M. and Warrier, A. M. (2004).
“Modelling of machinability parameters of carbon-carbon composite – a response
surface approach”, J. Mater. Process. Technol., Vol. 153-154, pp. 920-924.
[59] Gaitonde, V. N., Karnik, S. R., Rubio, J. C., Correia, A. E., Abrão, A. M., & Davim, J.
P. (2008). Analysis of parametric influence on delamination in high-speed drilling of
carbon fiber reinforced plastic composites. Journal of Materials Processing
Technology, 203(1–3), 431–438.
[60] Kishore, R. A., Tiwari, R., Rakesh, P. K., Singh, I., & Bhatnagar, N. (2011).
Investigation of drilling in fibre-reinforced plastics using response surface
methodology. Proceedings of the Institution of Mechanical Engineers, Part B:
Journal of Engineering Manufacture, 225(3), 453–457.
[61] Habib, M. A., Patwari, A. U., Jabed, A., & Bhuiyan, M. N. (2016). Optimization of
Surface Roughness in Drilling of GFRP Composite Using Harmony Search
Algorithm, 5(4), 311–316.
[62] Kilickap, E. (2016). Analysis and modeling of delamination factor in drilling glass
fiber reinforced plastic using response surface methodology. Journal of composite
materials, 45(6) 727–736
References
99
[63] Zadeh, L. A. “Fuzzy sets”, Information and control, Vol. 8, pp. 338-353, 1965.
[64] Karthikeyan, R., Jaiganesh, S. and Pai, B. C. (2002). “Optimization of Drilling
Characteristics for Al/SiCp Composites Using Fuzzy/GA”, Metals and Materials
Inter., Vol. 8, No. 2, pp. 163-168.
[65] Yaldiz, S., Unsacar, F. and Saglam, H. (2006). “Comparison of experimental results
obtained by designed dynamometer to fuzzy model for predicting cutting forces in
turning”, Mater. Design. Vol. 27, pp. 1139-1147.
[66] Latha and Senthilkumal (2009). “Fuzzy Rule Based Modeling of Drilling Parameters
for Delamination in Drilling GFRP Composites, Journal of Reinforced Plastics and
Composites 28(8):951-964
[67] Vimal, S. S. R., Latha, B. and Senthilkumar, V. S. (2009). “Modeling and Analysis of
Thrust Force and Torque in Drilling GFRP Composites by Multi-Facet Drill Using
Fuzzy Logic”, Intern. J. Recent Trends in Engg., Vol. 1, No. 5.
[68] Tsai, K. M. and Wang, P. J. (2001). “Comparisons of neural network models on
material removal rate in electrical discharge machining”, J. Mater. Proces. Technol.,
Vol. 117, pp. 111-124.
[69] Chinnam, R. B. and Baruah, P. (2004). “A neuro-fuzzy approach for estimating mean
residual life in condition-based maintenance systems”, Int. J. Mater. Product Technol.,
Vol. 20, Nos. 1-3, pp. 166-179.
[70] Samhouri, M. S. and Surgenor, B. W. (2005). “Surface roughness in grinding: On-line
prediction with adaptive neuro-fuzzy inference system”, Transactions of
NAMRI/SME, Vol. 33, pp. 57-64.
[71] Jagdev, S. and Simranpreet, S. G. (2009). “Fuzzy Modeling and Simulation of
Ultrasonic Drilling of Porcelain Ceramic with Hollow Stainless Steel Tools”, J.
Mater. Manuf. Proces., Vol. 24, No. 4, pp. 468-475.
[72] George, P. M., Raghunath, B., Manocha, L. M. and Warrier, A. M., (2004).“EDM
machining of carbon-carbon composite – a Taguchi approach”, J. Mater. Process.
Technol., Vol. 145, pp. 66-71.
[73] Xie, Y., Yu, H., Chen, J. and Ruan, X. (2007). “Application of grey relational analysis
in sheet metal forming for multi-response quality characteristics”, J. Zhejiang
University SCIENCE A, Vol. 8, No. 5, pp. 805-811.
[74] Chang, C. K. and Lu, H. S. (2007) “The optimal cutting-parameter selection of heavy
cutting process in side milling for SUS304 stainless steel”, Int. J. Adv. Manuf.
Technol., Vol. 34, pp. 440-447.
[75] Noorul Haq, A., Marimuthu, P. and Jeyapaul, R. (2008). “Multi response optimization
of machining parameters of drilling Al/SiC metal matrix composite using grey
References
100
relational analysis in the Taguchi method”, Int. J. Adv. Manuf. Technol., Vol. 37, pp.
250-255.
[76] Pal, S., Malvia, S. K., Pal, S. K., and Samantaray, A. K. (2009).“Optimization of
quality characteristics parameters in a pulsed metal inert gas welding process using
grey-based Taguchi method”, Int. J. Adv. Manuf. Technol., Vol. 44, pp. 1250-1260.
[77] Kurt, M., Bagei, E. and Kaynak, Y. (2009). “Application of Taguchi methods in the
optimization of cutting parameters for surface finish and hole diameter accuracy in
dry drilling processes”, Int. J. Adv. Manuf. Technol., Vol. 40, pp. 458-469.
[78] Jailani, H. S., Rajadurai, A., Mohan, B., Senthil kumar, A. and Sorankumar, T. (2009)
“Multi-response optimization of sintering parameters of Al-Si alloy/ fly ash composite
using Taguchi method and grey relational analysis”, Int. J. Adv. Manuf. Technol.,
Vol. 45, pp. 362-369.
[79] Ku, W. L., Hung, C. L., Lee, S. M. and Chow, H. M. (2010). “Optimization in thermal
friction drilling for SUS 304 stainless steel”, Int. Adv. Manuf. Technol., DOI
10.1007/s00170-010-2899-5.
[80] Lin, L. C., Lin, J. L. and Ko, T. C. (2002). “Optimization of the EDM Process Based
on the Orthogonal Array with Fuzzy Logic and Grey Relational Analysis Method”,
Int. J. Adv. Manuf. Technol., Vol. 19, pp. 271-277.
[81] Tosun, N. (2006). “Determination of optimum parameters for multiperformance
characteristics in drilling by using grey relational analysis”, Int. J. Adv. Manuf.
Technol., Vol. 28, pp. 450-455.
[82] Chiang, K. T., Liu, N. M. and Chou, C. C. (2008). “Machining parameters
optimization on the die casting process of magnesium alloy using the grey-based fuzzy
algorithm”, Int. J. Adv. Manuf. Technol., Vol. 38, pp. 229-237.
[83] Liu, D., Tang, Y., & Cong, W. L. (2012). “A review of mechanical drilling for
composite laminates”. Composite Structures, 94(4), 1265–1279.
[84] Montesano, John Bougherara, Habiba Fawaz, Zouheir (2017). “Influence of drilling
and abrasive water jet induced damage on the performance of carbon fabric / epoxy
plates with holes”. Composite Structures, vol. 163 pp. 257-266.
[85] Giasin, KhaledAyvar-soberanis, Sabino (2017). “3D Finite Element Modelling of
Cutting Forces in Drilling Fibre Metal Laminates and Experimental Hole Quality
Analysis”. Applied Composite Materials, pp. 113-137.
[86] Jasvir Singh and Ashwani Kumar (2018). “Original research paper experimental
study of drilling glass fiber reinforced epoxy resin composites ( GFREC ) ”. Indian
journal of research. 2 pp. 464-466.
List of Publications
101
List of Publications
Following papers are published/presented/under review at national/international
level journals/conferences.
1. Patel J.B: “Delamination free drilling of Glass Fiber Reinforced Plastics
(GFRP) and Carbon Fiber Reinforced Plastics (CFRP)-A review”, International
Journal of “Engineering, Science and Futuristic Technology 2015
(IJESFT2015), Volume 01 Issue 11, November-2015, ISSN: 2454-1125 having
impact factor of 2.94.
2. Patel J.B, Dr. M.B.Patel: “Effect of cutting parameters and tool geometry in
drilling of GFRP(Glass Fibre Reinforced Plastics)” , International Journal of
Advance Engineering and Research Development(IJAERD), Volume 05, Issue
06, June-2018, ISSN:2348-6406 having impact factor of 5.71
3. Patel J.B, Dr. Navneet Khanna: “ANFIS (Adaptive Neuro Fuzzy Inference
System) based model in MATLAB for selection of cutting parameters in
drilling of GFRP (Glass Fibre Reinforced Plastics)”, International Journal of
Precision engineering and Manufacturing (IJPEM), Springer Publications,
(Under Review).